In the past, producing a podcast meant choosing a topic, writing a script, inviting guests, recording audio, and spending hours on post-production editing.

For independent creators this was extremely time-consuming. For small teams with limited capacity — no dedicated person to record and produce episodes — podcasting often fell off the agenda altogether. As a result, many valuable ideas were never turned into podcast episodes.

AI changes that.

What Is an AI Podcast?

Simply reading an article aloud is nothing more than text-to-speech.

An AI podcast goes much further. You feed in source material — PDFs, webpages, interview transcripts, and other documents — and the AI first understands the content, identifies the key ideas, reorganises them into a natural conversational script, and finally converts that script into an engaging audio programme.

The most valuable part of the process is therefore not the synthesised voice itself, but the content organisation that happens beforehand. A well-made AI podcast sounds like two people discussing ideas they have already digested and understood, rather than a machine reading directly from a document.

Why Create AI Podcasts?

Research papers, technical blogs, course notes, and industry reports often stay locked in text, out of reach during a commute, a workout, or any other moment suited to audio. When the material piles up, reading all of it is hard to keep up with. Turning it into podcasts lowers the barrier to consuming information — you can listen instead of read.

AI podcasts also make content reuse much easier. A single document or research report can become notes, a blog article, and eventually a podcast, without rewriting everything from scratch each time.

And they dramatically reduce production costs, handling most of the repetitive work so you spend far less time recording, re-recording, and editing audio.

None of this is meant to replace traditional podcasts — it simply puts podcasting within reach when producing one the traditional way isn’t an option.

AI Podcast Platforms

Today there are already several mature AI podcast solutions available. For most use cases, third-party tools with well-designed workflows can already produce podcast episodes quickly and consistently.

NotebookLM

The best known and most capable of these is NotebookLM.

Its Audio Overview feature can automatically generate a podcast-style discussion from PDFs, webpages, video transcripts, and other uploaded materials.

You can access this feature through the NotebookLM web application.

Open the NotebookLM website at <https://notebooklm.google.com/> and click Create new notebook.

Paste a webpage URL into the dialogue box at the top, or upload downloaded documents into the NotebookLM workspace. In this example, we use Chapter 1 (pages 27–59) of Karl Marx’s Capital: A Critique of Political Economy, Volume I, published in 1867.

After the file is uploaded, NotebookLM automatically generates a summary, providing a quick overview of the document before you begin exploring it.

Hover over the Audio Overview button to see that NotebookLM can automatically transform the uploaded material into a podcast using AI.

Depending on the size of the source material, generation may take several minutes. For this 33 page excerpt, the process takes approximately 10 minutes.

Once generation is complete, the audio file appears in the lower-right corner. Click the play button to listen to the podcast.

NotebookLM also offers an Interactive Mode, making the listening experience much more engaging. Instead of passively listening, you can interrupt the AI hosts and ask questions at any point during the discussion.

Click Join to enter the conversation. The AI hosts pause the discussion, answer your question, and then seamlessly continue where they left off.

The podcast generated by NotebookLM carries a natural rhythm and clear delivery, weaving a paper’s core ideas into a fluid two-person dialogue. Compared to reading a lengthy research paper from start to finish, this format is far better suited to quickly absorbing the key takeaways during spare moments.

You can listen to the full podcast created by NotebookLM in the Bandung Circuits repository.

Open Source AI Podcast Generators

Compared to mature AI podcast platforms like NotebookLM, the greatest advantage of open-source projects lies in their higher degree of freedom and stronger controllability. Users can freely choose from different large language models according to their own needs, striking a better balance among generation quality, response speed, and cost.

At the same time, most open-source projects support local deployment, so data does not need to be uploaded to third-party platforms, making them more suitable for scenarios that require privacy, security, or corporate data compliance. In addition, such projects usually support custom podcast personas, automation workflows through APIs, and allow developers to extend features or modify source code based on actual needs, unrestricted by commercial platforms.

If you want to delve deeper into AI podcasts, carry out secondary development, or build your own AI podcasting system, open-source solutions undoubtedly offer greater playability and room for expansion.

ai-podcast-creation Skill

The simplest open-source option is the ai-podcast-creation skill. Unlike NotebookLM, which produces a finished audio podcast, it generates the podcast script only. You can then bring that script to life however you prefer — record it yourselves, or convert it to audio with a text-to-speech tool. This suits a team whose hosts still want to voice the episode but would rather save the time writing the script.

Installing the skill is straightforward. Open Agent and enter the following command:

Please help me install this skill:
npx skills add inference-sh/skills@ai-podcast-creation

After installation finishes, restart VS Code to load the skill into your session. You can then generate a podcast script using a prompt such as:

/ai-podcast-creation Please generate a podcast for @project/AI-podcast/Capital-Volume-I-pages-27-59.pdf

The primary purpose of this skill is to generate podcast scripts, with the final output presented as a natural conversation between multiple hosts.

The full script is available in the Bandung Circuits repository.

Open Notebook

Open Notebook has a very clear positioning: it aims to become the open-source alternative to NotebookLM.

Officially, it is defined as an AI platform for research, learning, and knowledge management, not merely a podcast generation tool. As a result, besides supporting AI podcast generation, it also offers common NotebookLM features such as AI Q&A, AI notes, and AI summarisation.

More importantly, Open Notebook is also very friendly to ordinary users. The project provides a complete Web UI, so even without any programming background, you can generate AI podcasts through a visual interface, just like using ordinary desktop software.

Before installing it, there is one prerequisite worth understanding: because Open Notebook runs on your own computer but still relies on external AI models, you will need an API key.

What Is an API Key?

If you plan to try out local AI projects, an API key is essentially unavoidable. Although many AI applications run locally, the core capabilities that actually consume computing resources and determine output quality usually still come from third-party AI services.

Take AI podcasting as an example: the entire generation pipeline typically involves multiple steps such as article comprehension, content summarisation, script generation, and speech synthesis. If you rely entirely on local models to complete these tasks, not only will you need a fairly high hardware configuration, but the generation speed and final quality will often fall short compared to cloud-based models, making the overall cost-effectiveness quite low.

Therefore, most projects will call model services such as OpenAI, Google, Anthropic to handle these core tasks, and an API key is essentially the key to accessing these services. Only after configuring an API key in your local project can you successfully invoke the corresponding models. Of course, model providers will charge API fees based on actual usage.

Installing and Configuring Open Notebook

It is recommended to use an AI agent to automatically complete the installation and configuration of Open Notebook. After opening VS Code, simply enter the prompt below.

Set up Open Notebook on this computer.

If Git is installed, clone https://github.com/lfnovo/open-notebook.git. Otherwise, download https://github.com/lfnovo/open-notebook/archive/refs/tags/v1.10.0.zip and extract it.
Follow the instructions in README.md to install dependencies and start the application.

After the installation completes, the Web UI will usually open automatically. If it does not start automatically, you can visit http://localhost:8502 in your browser.

Click the Models button in the lower left corner to enter the model configuration page.

Open Notebook supports multiple model service providers. Below we will use OpenAI as an example for configuration.

First, go to the OpenAI platform to create an API Key.

https://platform.openai.com/api-keys

Note that the OpenAI API and ChatGPT Plus subscription are not interchangeable; the API service requires a preloaded balance before it can be used normally.

https://platform.openai.com/settings/organization/billing/overview

Click Create new secret key to create a new API Key.

Fill in the API Key name, then click Create secret key.

The API Key will be shown only once, so it is recommended to copy and save it immediately. Its format is usually sk-xxxxx.

Then return to Open Notebook and click Add Configuration.

Fill in the Configuration Name and paste the API Key you just saved to complete the setup.

You can click Test to check whether the API can be called successfully. If the test fails, first verify that your API account balance is sufficient.

After confirming the configuration is successful, click Models to start adding models.

Open Notebook requires you to configure four types of models:

• Language (Large Language Model): Responsible for text tasks such as conversation, summarisation, reasoning, and content generation.
• Embedding (Text Embedding Model): Responsible for converting text into vector embeddings, used for RAG, semantic search, similarity computation, and related functions.
• TTS (Text-to-Speech): Responsible for converting text into speech.
• STT (Speech-to-Text): Responsible for speech recognition and transcription.

Below that, all models currently supported by OpenAI will be displayed.

If you are unfamiliar with the differences between these models, you can hand the model list over to AI and let it help recommend the most suitable configuration. (Replace [Model List] below with the actual model list.)

I need to select the most appropriate model for each of the following four task categories from the model list below, and I would like you to explain your recommendations.

The task categories are:

**Language (LLM):** General-purpose text generation, including conversational AI, question answering, reasoning, summarization, and content generation.

**Embedding:** Converting text into vector embeddings for semantic search, retrieval-augmented generation (RAG), similarity search, clustering, and related applications.

**TTS (Text-to-Speech):** Converting text into natural-sounding speech.

**STT (Speech-to-Text):** Transcribing spoken audio into text.

For each task category, please:

- Recommend the best-suited model.
- Briefly explain why it is recommended, considering factors such as capability, performance, latency, and cost.
- Clearly indicate if no suitable model exists for a particular category.
- If any listed model is deprecated or outdated (for example, `text-embedding-ada-002`), recommend the most appropriate modern replacement instead.

Using the following model list:
[Model List]

In this article, we will use the model combination recommended by GPT to generate a podcast:

• Language: gpt-5.5
• Embedding: text-embedding-3-large
• TTS: gpt-audio
• STT: gpt-4o-transcribe

For example, set Model Type to Language, choose gpt-5.5, then click Add to add the model.

After adding, you can select gpt-5.5 from the Chat Model dropdown above.

Follow the same method to add all the models required by Open Notebook one by one.

Then go to Podcasts → Profiles to configure the corresponding models for your podcast presets.

Here we take tech_discussion as an example; click Edit.

For all configuration items marked with *, select the models you have already added, such as gpt-5.5.

After completing the configuration, check whether the right side still shows Setup required. If this reminder disappears, it means the current template has been fully configured.

Continue configuring the remaining templates until all Setup required indicators have disappeared; then you can officially start generating podcasts.

Creating a Notebook and Importing Articles

First, click New → Notebook to create a new notebook.

Enter the notebook name and click Create New Notebook.

After creation, you will see the new notebook on the Notebooks page on the left.

Enter the notebook and click Add Source to import the article you want to convert into a podcast.

Open Notebook supports three import methods: entering a web page URL, uploading a local file, and pasting text directly. This article demonstrates the second method by uploading a local file.

After the system finishes parsing, the imported article will appear in the Sources section, indicating that the material has been successfully indexed.

Generating an AI Podcast

After importing the materials, click New → Podcast again to create a new podcast project.

The Episode Settings on the right allows you to set the podcast style. Open Notebook provides three default presets:

• business_analysis: A business analysis, financial interpretation, and case study style, generally more formal.
• solo_expert: A solo expert presentation, similar to a lecture or a teacher’s lesson.
• tech_discussion: A two-host discussion on technology, papers, AI, programming, and similar topics, closer to NotebookLM’s podcast style.

In addition to the presets, you can add extra instructions in Additional instructions, such as requesting more humour, adjusting speech rate, controlling the episode length, or specifying the style of expression.

After confirming the settings, click Generate, and Open Notebook will start generating the podcast.

Checking Generation Progress

After the generation task is submitted, click the Podcasts page on the left to view the current task list.

Depending on the article length and model response speed, the entire generation process usually takes a few minutes. When the task is completed, the corresponding podcast will appear in the list.

After the podcast conversion is complete, click the play button at the bottom of the page to listen to the AI podcast generated by Open Notebook online.

If you are not satisfied with the podcast content, host style, or delivery, you can return to Episode Settings, adjust the preset template or modify the Additional instructions, and then regenerate.

Since Open Notebook supports freely choosing large language models and speech models, swapping different model combinations can also yield vastly different podcast results. For users who want to build a personalised AI podcast workflow, this high degree of configurability is one of the greatest advantages of the open-source approach.

You can listen to the full podcast created by Open Notebook in the Bandung Circuits repository.

Conclusion

AI has fundamentally changed the podcast production workflow. Instead of spending most of your time on scripting, recording, and editing, you can now begin with existing documents, allow AI to understand and reorganise the material, generate a conversational script, and finally produce a polished audio programme.

For most creators, the greatest value is not the synthesised voice itself, but AI’s ability to transform complex written material into engaging conversations. Whether you are working with research papers, technical documentation, course materials, or industry reports, AI makes it possible to repurpose knowledge into an accessible audio format with far less effort than traditional podcast production.

Raw transcripts are often only the starting point of content organisation. Whether from meetings, interviews, lectures, or internal discussions, speech-to-text output usually contains spoken-language expressions, repetitions, incomplete sentences, and occasional recognition errors, making it unsuitable for direct reading.

Traditionally, organising these materials required substantial manual effort. Today, large language models can assist with tasks such as extracting key points, improving structure, and refining content.

After transcription, there are generally two common output formats: meeting minutes and formal articles.

Meeting minutes are a faithful record. Their job is to capture discussions, viewpoints, decisions, and action items completely and in the meeting’s own order, so that anyone can trace who said what and why a decision was reached. They are typically itemised and meant to be scanned or searched rather than read straight through.

A formal article, by contrast, is a piece of writing in its own right. Rather than recording the discussion, it selects only the most valuable insights and reorganises them by theme into a continuous, self-contained argument. It can be read and understood without having attended the meeting, and it favours depth, structure, and narrative flow over sheer completeness.

Therefore, before processing a transcript, it is important to determine the intended output. The goal will directly affect both the prompting strategy and the organisation process.

All example outputs in this guide are generated from the same transcript — a recording of the Hands Off Asia webinar held on 30 April 2025.

Generating Meeting Minutes

Meeting minutes are common in business, project management, and academic settings. Their primary purpose is to help readers quickly understand what happened, what issues were discussed, what viewpoints were raised, and what decisions or follow-up actions were agreed upon.

Important Considerations

In practice, the most common problem when generating meeting minutes is not a lack of fluency but omitted information.

Many users will simply copy a transcript tens of thousands of words long into the AI and type a brief instruction like:

Please summarise this meeting.

For short content, this approach may yield decent results. However, when the meeting content is long, the model may overlook certain parts or even hallucinate due to the excessive context length, adding information that was not present in the original text.

A more recommended method is to process the transcript in stages. First, have the model summarise the content chapter by chapter or by time segment, and then merge these summaries into a complete set of minutes. Compared to compressing the full text in one go, this method is generally more stable and makes it easier to spot any omissions.

For large meetings that span several hours, you can even adopt a “three-step method”: first generate segment summaries, then consolidate all summaries, and finally produce the formal meeting minutes. Although this involves more steps, the result tends to be much more reliable.

Using the Web Version of AI to Generate Meeting Minutes

For most users, the simplest approach remains using the web version of AI directly.

Current mainstream models like ChatGPT, Claude, Gemini, and Qwen all support uploading TXT, DOCX, or PDF files. After uploading the transcript, you can simply ask the model to generate the meeting minutes.

To achieve more consistent results, it is advisable to clearly state in the prompt what the minutes should include, such as the meeting background, discussion topics, decisions made, and action items.

An example prompt is as follows:

Please convert this transcript into professional meeting minutes.

Include:

- Meeting overview
- Key discussion topics
- Decisions made
- Action items
- Open questions

Do not omit important information.

Do not invent content.

Keep the logical order of the original meeting.

Compared to using only the brief instruction Please summarise this meeting, this prompt not only ensures that the model accurately retains the meeting’s core information and logical sequence but also automatically completes the structural reorganisation from a word-for-word record to professional meeting minutes. The output is strictly organised according to a standard minutes framework, including Meeting Overview, Key Discussion Topics, Decisions Made, Action Items, and Open Questions, transforming the originally lengthy and scattered discussion into a document that is easy to read, search, and use for subsequent execution.

The complete output is available in the Bandung Circuit repository.

Beyond this, you can further adjust the prompt according to your specific needs. For example, for policy research, media reports, or institutional archiving, you can request a summary version that emphasises core arguments and main conclusions. For academic exchanges, international forums, or specialised seminars, you can retain more of the speakers’ backgrounds, lines of reasoning, and the different participants’ viewpoints to facilitate subsequent research, citation, or content organisation.

When the transcript is particularly long, it is recommended to process it in batches according to the agenda, chapters, or time periods. This not only helps reduce the risk of information loss that can occur when handling very long texts, but also preserves the contextual relationships within each discussion section more accurately, making it easier to verify, supplement, and integrate later, ultimately producing a more complete and accurate set of meeting minutes or event records.

Using a Skill to Generate Meeting Minutes

For users already employing skills within an Agent or Visual Studio Code, the task of generating meeting minutes can be further automated. One commonly used option is the meeting-minutes skill.

This type of skill is usually optimised specifically for meeting scenarios. Besides summarising the discussion, it can also automatically extract decisions, action items, and follow-up tasks.

Installing this skill in an Agent is straightforward. Open the Agent and enter:

Please help me install this skill:

npx skills add https://github.com/github/awesome-copilot --skill meeting-minutes

After the installation is complete, restart Visual Studio Code to refresh the skills list, and you can then invoke it directly. For example:

/meeting-minutes Generate professional meeting minutes from @meeting_transcript.srt

For project meetings, research discussions, and planning sessions, the value of such a specialised skill lies not only in more consistent results but in its built-in information extraction rules tailored to meeting contexts. Compared to standard meeting minutes generated via prompts, a skill will often go further to identify attendees, the meeting agenda, decisions, action plans, risk factors, and follow-up items, organising everything according to a fixed template.

As you can see from the output, the content generated by the skill is no longer just a record of the meeting; it resembles a complete meeting management document. For example, in addition to the usual Meeting Overview, Discussion Topics, and Action Items, it also produces sections like Metadata, Attendance, Agenda, Risks / Blockers, Next Meeting, and Attachments, and supplements each task with an owner, acceptance criteria, and related resource links. This format is much better suited for team collaboration, project tracking, and knowledge archiving, rather than simply documenting what was discussed.

The complete output is available in the Bandung Circuit repository.

Turning Transcripts into Formal Articles

Turning a transcript into an article is not simply a matter of shortening it. A discussion unfolds in the order people happen to speak: arguments are scattered across different voices, points are repeated, and ideas often arrive half-formed before being picked up again later. Writing it up as an article means reshaping that raw material — drawing out the core arguments, grouping related points by theme, cutting repetition and filler, and rebuilding them into a single line of reasoning a reader can follow from beginning to end. The harder part is doing all this while staying faithful to what the speakers actually meant: their terminology, their emphasis, and their stance.

Prompting the AI to Write the Article

1. Defining the Editor Identity

The first step in generating an article is not telling the AI to write, but telling it what identity it should write from. Without a role definition, models tend to default to a generic writing mode: lightly rephrasing sentences, compressing repetitive content, and mechanically organising paragraphs. Actual editing, however, reconstructs spoken material into a complete written argument. You therefore need to explicitly tell the model that it is not a summarisation assistant or a meeting minutes tool, but a professional editor.

## Your Role


You are an experienced editor specialising in transforming raw transcripts into publication-ready articles.

Your task is not to summarise the transcript or produce meeting notes. Instead, you must restructure spoken content into a coherent written argument suitable for public readers while remaining faithful to the original speakers' meaning, terminology, and intent.

The final article should read as a polished piece of analytical writing rather than an edited transcript.

2. Identifying the Content Type

There is no single way to rewrite a transcript. A multi‑speaker panel discussion and a solo lecture have fundamentally different content structures. If the content type is not identified first, the AI can easily apply the wrong organisational approach. Before formal writing begins, therefore, the model should complete a content classification.

## Identify the Transcript Genre


Before writing, determine the primary format of the transcript.

Possible genres include:

- Panel Discussion
- Webinar
- Lecture
- Keynote Speech
- Interview
- Conference Session
- Debate
- Roundtable Discussion

Select the dominant genre and choose an appropriate restructuring strategy.

For example:

- Multi-speaker discussions should be reorganised around themes.
- Lectures should preserve the original argumentative sequence.
- Interviews may remain in Q&A form or be converted into a feature article.
- Debates should preserve disagreements and opposing positions.

The purpose of this step is to make the AI decide on the article’s organising logic before it begins writing.

3. Outlining First Instead of Writing Directly

Many people ask the AI to output a complete article right away. In practice, the most important step in editing is often not writing but planning the structure. A good outline determines the final quality of the article. It is therefore advisable to split the generation process into two phases: plan first, then write.

## Outline First

Before drafting the article, generate a detailed outline.

The outline should include:

- Proposed article title
- Main thesis
- Section headings
- Key arguments under each section
- Possible opening angles

Also identify any information that may require verification, including:

- Personal names
- Organisations
- Locations
- Dates
- Statistics
- Quotations

Do not guess uncertain information.

This approach lets you check whether the structure makes sense before moving into the formal writing stage.

4. Establishing Fidelity Rules

The greatest risk in rewriting transcripts is not language quality but factual errors. When organising content, models tend to automatically supply background knowledge, infer missing information, and even construct logical chains that did not originally exist. Clear source boundaries must therefore be set.

## Source Fidelity Requirements

The article must be based exclusively on the materials provided by the user.

Use only:

- The transcript
- Supporting documents supplied by the user

Do not:

- Search the web
- Add external context
- Invent facts
- Invent quotations
- Introduce unsupported claims

If information is uncertain, flag it instead of guessing.

This section is effectively the most important safety mechanism in the entire skill.

5. Preserving the Original Stance and Expression

Many models exhibit a common tendency: they automatically neutralise positions. Sharp political, social, or academic expressions are rewritten into milder, vaguer language. For news commentary, research institute articles, or political interviews, this often distorts the original meaning. The model therefore needs to be explicitly told to retain the original expression.

## Preserve the Original Voice

Maintain the speakers' terminology, framing, and political language.
Do not soften or neutralise terms used in the source material.

Preserve key concepts exactly as they appear whenever possible.

Avoid introducing artificial balance or alternative viewpoints that do not exist in the source.

The goal of this section is to preserve the author’s thinking, not merely the facts.

6. Specifying the Article Style

Once the content structure and factual boundaries are determined, the final step is style control. This part tells the AI what the finished piece should resemble.

## Writing Style

Write in the style of serious analytical non-fiction.

The article should:

- Read like a research institute publication or long-form commentary.
- Present ideas directly rather than reporting who spoke first.
- Develop a coherent argument.
- Use thematic sections with meaningful headings.
- Maintain a formal and professional tone.
- Conclude with implications, significance, or consequences.

Avoid sounding like:

- Meeting minutes
- Event reports
- Raw transcripts
- Generic AI summaries

Through this layer of constraint, the model shifts from “organising content” to “constructing an article.”

7. Output Format

Finally, the format of the deliverable needs to be clearly specified.

## Output Format

Produce the article as a single Markdown document.

Structure:

# Title

Introduction

## Section One

Content

## Section Two

Content

## Conclusion

Content

Output only the final article in Markdown format.

8. Input Content

Once all the rules above have been defined, the last step is simply to place the transcript in the input area. If you have any supporting materials — background documents, speaker notes, or reference texts — you can include them here too, so the model can draw on them while still staying within the source boundaries set earlier.

## Input Transcript
[Paste the full transcript here.]

## Supporting Materials (optional)
[Paste or attach any background documents, speaker notes, or reference texts here.]

Running this prompt on the transcript produces an article that has completely shed the form of meeting minutes. Content originally presented by different speakers is re‑integrated into a continuous discussion centred on issues such as sovereignty, security, militarisation, and regional peace. This style of writing is closer to academic commentary, thematic analysis, or in‑depth media features:

The complete output is available in the Bandung Circuit repository.

Using a Skill to Generate Articles

Beyond writing prompts directly, a purpose‑built skill can also be used within an Agent workflow to convert transcripts into formal articles. Compared with one‑off prompts, such skills typically come with a more complete rule system, can handle different types of transcript content more consistently, and produce articles with a clear structure, coherent logic, and a style suitable for public reading.

Global South Insights has developed and open-sourced a more fully featured transcript-to-article skill for turning raw transcripts such as meeting notes, interviews, and lecture recordings into high‑quality articles. The skill has built‑in capabilities for article restructuring, logical organisation, language polishing, and style unification, significantly raising the quality of the final output. You can obtain and install this skill from the corresponding repository.

The installation process is very simple. After downloading the skill, just tell the Agent:

Please help me install this skill

The Agent will automatically complete the installation and configuration.

Once installed, you can directly process a transcript file. For example:

/transcript-to-article Create a detailed article outline from @transcript.txt

After execution, the Agent will first analyse the transcript, identify the thematic structure and core arguments, and automatically generate a detailed outline, providing a clear framework for subsequent article writing.

The real difference between this skill and an ordinary prompt is not the quality of the prose but how much of the process it manages for you. A one-off prompt does everything in a single pass and leaves every rule for you to specify. The skill instead runs the job in stages — it produces an outline for your approval before writing any prose — and adds controls a single prompt cannot easily reproduce: it flags the names, dates, and figures it is unsure the transcription got right (drawing on any supporting materials you provide), detects the transcript’s genre and restructures accordingly, and writes to a defined house style you can steer with a sample article, a target length, and a glossary.

The result shows that the article does not simply unfold by country or speaker. Instead, it advances layer by layer around core issues such as “the architecture of militarisation,” “military bases and the security paradox,” “soft power and economic control,” “the price borne by ordinary people,” and “transnational resistance and international solidarity.” Points originally scattered across multiple speakers’ remarks are integrated into a unified chain of argument, making the article closer to a media special report, opinion piece, or an analytical report issued by a research institute.

In addition, this skill automatically generates an HTML presentation page that lays out the article’s structure, key points, and images more clearly, making the piece easier to read, present, and publish online.

The complete output is available in the Bandung Circuit repository.

Turning Transcripts into PowerPoint Presentations

If you need to take meeting minutes, interview records, course content, or seminar transcripts and further use them for sharing and reporting, you might consider organising them into a PowerPoint presentation. Compared to lengthy text, a presentation is more suitable for showcasing core ideas, clarifying the logical structure, and helping the audience quickly grasp the main points.

In our How to Guides series, we have previously covered how to use AI to organise research notes and textual materials into a slide deck. For more details, please refer to How to Present Your Research — From Notes to Slides.

Conclusion

A transcript is essentially raw material, not the finished product. Whether it is a meeting discussion, an academic interview, or a course recording, it always requires further organisation to truly unlock its value.

If your goal is to document the meeting process and preserve the discussion content and decisions, then meeting minutes are usually the more appropriate choice. If your goal is to communicate ideas, write a report, or publish publicly, you need to further distill the content and restructure it according to the readers’ needs.

As the capabilities of large language models continue to improve, a relatively complete workflow has emerged, spanning from audio transcription to meeting minutes and finally to formal article generation. With well-designed prompts and a layer of human review, a single recording can now become usable minutes or a publishable article in a fraction of the time the work once took — cutting the human effort and cost of turning a meeting into something worth reading, and freeing people to focus on the ideas rather than the write-up.

In academic research, meeting minutes, interview collation, and video subtitle production, converting audio or video content into editable text quickly is one of the most common and fundamental data organisation tasks. Compared with manual dictation, using speech recognition tools is not only more efficient, but also more convenient for subsequent proofreading, retrieval, and archiving.

This article will introduce how to transcribe interviews, meetings, and other audio and video materials into text using several common tools, while preserving a relatively complete graphical operation workflow for practical use.

The example recording used in this guide is a guest presentation at the Global South Academic Forum.

Transcribing with Standalone Software

Standalone software is usually the most beginner-friendly approach: you download and install a desktop application, then complete the entire workflow — from upload to export — within a single interface. The tool we recommend here is iFLYREC.

It provides a relatively complete graphical transcription solution, integrating speech recognition, speaker diarisation, domain-specific optimisation, and result export on the same platform. For users who want to avoid script configuration and still require high transcription efficiency, this kind of tool is very suitable for daily office work and research organisation.

Its advantage lies in its clear workflow: users only need to upload a file, select language and scene parameters, and the system will automatically generate a preliminary text. Afterwards, proofreading can be carried out with the help of audio-text synchronised playback, timestamp positioning, and speaker labelling, thereby reducing manual rework. The following uses the transcription of audio and video files as an example to illustrate its basic usage.

1. Software Installation and Account Login

First, visit the official iFLYREC download page:

https://www.iflyrec.com/zhuanwenzi.html

After entering the page, click ‘Download’ to obtain the latest client installation program.

Once the download is complete, double-click the installation package to start the installer and follow the installation wizard to complete the software deployment step by step.

When launching the software for the first time, you need to read and agree to the user agreement, and then log in to the system using your iFLYTEK account.

After logging in, you can enter the software’s main interface and start the subsequent transcription tasks.

2. Overview of the Main Interface

The iFLYREC homepage integrates various speech-to-text capabilities, covering usage needs in different scenarios. The interface provides relatively direct entry points for tasks such as real-time recording, existing file transcription, and subtitle creation, and the usage logic is relatively clear.

The Recent Files area at the bottom of the software displays recently processed tasks, allowing users to review historical files at any time. Once tasks are synchronised to the cloud, it is also convenient to continue processing on different devices, reducing the burden of local file management.

In practical applications, the most common need is not real-time recording, but converting already recorded audio and video materials into text. Therefore, this article focuses on the Import File function.

3. File Upload and Parameter Configuration

Transcription quality is not only closely related to the clarity of the audio itself but also to the initial parameter settings. Before uploading a file, setting options like language, speaker number, and professional domain can usually make the recognition results more stable and simplify subsequent proofreading.

3.1 Selecting the Source Language

The source language setting is an important factor affecting recognition accuracy. Users should choose a language type that matches the audio content as closely as possible, such as English, Chinese-English mixed, or Chinese. When the language is selected incorrectly, the system often produces more errors, especially in cases involving proper nouns and complex long sentence structures.

Clicking More allows you to view a wider range of language options, including Spanish, Japanese, Chinese, and many other languages, as well as enhanced models like English Pro and Chinese-English Mixed Pro.

In actual use, choosing the right language model accurately is often more important than repeatedly making corrections later. For content with many specialised terms, non-standard pronunciation, or mixed-language expressions, selecting an appropriate model can significantly improve readability.

3.2 Importing Audio/Video Files

After completing the language selection, you can drag the file to be processed directly into the upload area, or import it through the file selection window.

It supports a wide range of video and audio formats, covering the vast majority of meeting recordings, course videos, and interview materials. For longer materials, it is recommended to confirm the file naming and source first, so that they are easier to identify during subsequent organisation and export.

3.3 Setting the Number of Speakers

In multi-person interviews or meetings, distinguishing between different speakers is very important for subsequent analysis. The clearer the speaker diarisation, the easier the final text is to read and the more convenient it is for topic summarisation and viewpoint comparison during research.

iFLYREC offers a speaker diarisation feature, allowing users to manually specify the number of speakers or select Auto to let the system determine it automatically.

For example, if the file is a single-person speech, course recording, or podcast interview, you can directly select 1 Speaker. For multi-person discussion scenarios, it is recommended to set the number of people as accurately as possible to reduce errors in the system’s allocation of dialogue turns. More accurate speaker information typically leads to a clearer and more standardised transcription result.

3.4 Configuring Professional Domains

For specialised scenarios such as law, economics, and medicine, iFLYREC also provides domain-specific optimisation models.

When the transcription content involves many specialised terms, selecting the corresponding domain can often effectively reduce the term recognition error rate and improve the overall readability of the text. For instance, if company names, research jargon, or industry abbreviations frequently appear in a meeting, the system is more likely to make reasonable judgments if it knows the general domain.

If the system’s provided domain categories cannot fully match your research topic, you can also use the Keyword Optimization feature to customise keywords for auxiliary optimisation.

This type of setting is particularly suitable for academic interviews, industry exchanges, and thematic discussions. The system will combine keyword information to make more targeted corrections to the recognition results, bringing the final text closer to the original semantics.

3.5 Submitting the Transcription Task

After completing the parameter settings, click Submit to upload the file and submit the task.

During the upload process, the system will display real-time progress. Once the upload is complete, the original Uploading status will change to Open File.

Click it to enter the transcription interface and start processing the task. This process may take some time for large files or long recordings, but the overall operation logic remains relatively intuitive.

4. Viewing Transcription Results and Quality Proofreading

While a task is running or after transcription is complete, the system will display the corresponding file in the left taskbar.

Users can open the task at any time to view the recognition results. iFLYREC uses an audio-text linkage method for proofreading: when you click on any paragraph in the text, the player will automatically jump to the corresponding time position.

This design is very practical for collating interview materials, verifying meeting minutes, and performing sentence-by-sentence proofreading in academic research. It helps users quickly confirm who said a particular sentence, whether the original speech was recognised correctly, and whether the context segmentation is reasonable.

4.1 Preview Settings

The settings menu next to the player offers several auxiliary reading functions.

Among them, Display Speaker shows identity tags for different speakers; Speaker Filtering can filter playback content by speaker; Display Timecode shows the timestamps corresponding to text paragraphs; and Skip Silent Segments can automatically skip silent parts, making reading and playback more compact.

These settings are particularly helpful when dealing with long meetings, in-depth interviews, and classroom recordings. For users who need to quickly locate content, they not only improve efficiency but also reduce the hassle of repeatedly dragging the progress bar.

4.2 Automatic Translation Function

In addition to speech-to-text, iFLYREC also provides full-text translation capabilities. Clicking the translation button in the interface will perform an automatic translation of the transcribed text.

For cross-language interviews, international meetings, and foreign language course collation, this function can significantly reduce the manual translation workload and makes it convenient to obtain a readable Chinese version first for a second round of revision.

It should be noted that iFLYREC primarily supports translating multiple languages into Chinese. If you need to generate versions in other target languages, it is recommended to process them further with professional translation tools after export to ensure accuracy of expression.

5. Exporting and Saving Results

After completing the proofreading, you can go to the Downloads page to export the final results.

The system supports multiple output formats, including DOCX (Word Document), TXT (Plain Text), and SRT (Subtitle File). You can also decide whether to retain metadata such as timecodes and speaker information as needed.

After confirming the parameters, click Download to save the file to your local device.

For video post-production, the SRT format is usually the most commonly used subtitle output format and can be directly imported into most editing software for subsequent editing.

From a practical application perspective, iFLYREC can well meet the needs of scenarios such as meeting recording, interview collation, lecture transcription, and video subtitle production. Its advantage lies not only in its high recognition efficiency but also in its integration of speech recognition, speaker diarisation, professional term optimisation, translation, and export into a complete workflow, lowering the barrier for non-technical users to carry out speech-to-text tasks.

For researchers, media workers, and content creators who frequently need to process large amounts of audio and video materials, reasonably configuring the language, speaker, and domain parameters, and combining this with a post-processing proofreading workflow, usually allows them to quickly obtain high-quality, structured text results.

Transcribing with Web-Based AI Tools

In addition to standalone transcription software, some AI web interfaces also provide online recording and meeting minutes functions. Qwen’s meeting minutes feature falls into this category. It can record content in real-time during a meeting, categorise speech based on different speakers, and finally generate a relatively complete text summary.

After entering the Qwen official website at https://www.qianwen.com/chat, click ‘More’ → ‘Meeting Minutes’ to enter the recording page.

Once on the meeting minutes page, you need to grant microphone permission so that the system can record audio normally.

During the recording process, the system not only records the speaking time but also distinguishes between different speakers based on timbre and voiceprint characteristics, making it quite practical for multi-person discussion scenarios.

After clicking stop, Qwen will automatically summarise the content and generate a meeting report.

This approach is suitable for impromptu meetings, online discussions, and quick minute-taking, especially for users who don’t want to install extra software. Its advantage lies in its quick learning curve and short workflow, making it ideal for immediate use.

Transcribing with a Local Agent Skill

In addition to cloud-based online tools, you can also complete audio transcription by configuring skills in a local Agent. For example, OpenAI’s Whisper skill is a relatively common type of transcription tool that can run locally and is suitable for users with higher requirements for privacy and workflow.

The method for installing the Whisper skill in an Agent is also quite straightforward. Open the Agent and type:

Please help me install this skill:

npx skills add steipete/clawdis@openai-whisper

After waiting for the installation to complete, restart VS Code to refresh the skills list. Then, when you need to transcribe an audio or video file, simply input:

/openai-whisper Transcribe the audio/video file @file_to_transcribe.mp4 and output the result as a plain text (.txt) transcript.

If you need a subtitle file instead, you can also change the output format to .srt, thus obtaining standard subtitle text suitable for video post-production use.

The characteristic of this method is its greater controllability, making it suitable for users who have clear requirements for output format, file path, and processing workflow. However, it also depends more heavily on local environment configuration, making it better suited for people already using an Agent workflow.

Transcribing with Cloud APIs

If you prefer not to transcribe audio locally, you can also directly use the API services provided by major model vendors. The advantage of this approach is that it avoids the computer overheating, performance usage, and memory pressure caused by running a local model for a long time, making it especially suitable for scenarios with long audio files and many tasks.

In fact, most common AI model providers on the market offer audio transcription capabilities, such as OpenAI, Gemini, Claude, and the previously mentioned Qwen. Users only need to apply for an API Key on the corresponding platform, and their local Agent can call cloud capabilities to complete the transcription. However, these services are typically paid.

Take OpenAI’s API Key as an example below. OpenAI’s API Key can be obtained at https://platform.openai.com/api-keys and usually starts with sk-.

After copying the API Key, you can save it in a file or write it into an environment variable for subsequent calls. Then, you can reference this key in the Agent, for example:

Use the OpenAI API key provided in @openai_api_key.txt to transcribe the audio/video file @file_to_transcribe.mp4. Output the result as a plain text (.txt) transcript.

The Agent will call OpenAI’s audio transcription model to convert the video or audio content into text. Generally, the results obtained this way are of high quality, especially suitable for materials that are clear, have formal content, and are long in duration. The usage for other model vendors is essentially similar; as long as you obtain the corresponding API Key, you can integrate the transcription capability into your own workflow.

Summary

This article introduced several common audio-to-text methods, including iFLYREC, Qwen meeting minutes, local skill transcription, and cloud API calls. Different methods have their own focuses: graphical tools are more suitable for regular users to get started quickly, local skills are better for users who already have an Agent workflow, and the API solution is more suitable for scenarios requiring stable, batch processing of audio and video materials.

Regardless of the method used, the actual effect usually depends on three key factors: audio quality, parameter settings, and post-processing proofreading. As long as you clarify the language, speaker, and professional domain as much as possible before transcription, and make necessary revisions before exporting, you can quickly obtain high-quality text that can be used for research, organisation, and publication.

AI Is Transforming Research Presentations

In the past, creating a research presentation often meant spending a great deal of time manually organising content, adjusting layouts, and repeatedly revising PowerPoint slides.

With the rapid development of large language models and generative AI, research presentations are entering a new stage. Today, research notes, paper summaries, and experimental results can first be organised by AI and then further turned into slides, infographics, or web-based presentations. For progressive research institutions in the Global South, however, commercial tools pose two distinct problems: most are designed for business reporting rather than academic expression, technical content, or structured visualisation; and licensing costs put many of them out of reach for institutions operating under resource constraints. Open-source projects offer an alternative on both fronts — better suited to research work, and accessible regardless of institutional budget.

The following sections introduce several open-source presentation projects that work with large language models to help researchers generate presentations that are more polished and structurally clearer.

HTML Presentation Generation Tools

Generating a Deck End-to-End with an AI Skill

One approach to AI presentation generation is to use a skill that handles the full pipeline — from outline to finished slide deck — through an HTML-first workflow. This means the slides are essentially webpages rather than traditional PPT files, which is especially well-suited to research scenarios: large language models are naturally better at handling structured text than complex Office files. Researchers can use Markdown, HTML, or component-based structures to organise their content, and then convert that content into presentation pages. The advantages are clear: greater layout flexibility, clearer structure, and easier integration of dynamic charts, code blocks, and interactive content.

This kind of skill is also highly suitable for collaboration with AI. Researchers can first ask the model to generate a presentation outline, and then automatically convert it into slides, saving a large amount of repetitive work.

One such skill is Frontend Slides, among the more capable options currently available in the open-source ecosystem. It can be installed directly through the plugin marketplace. Open Agent and enter:

Please help me install these plugins:

/plugin marketplace add zarazhangrui/frontend-slides
/plugin install frontend-slides@frontend-slides

After installation, you may need to restart Agent once to refresh the plugin. You can then use the /frontend-slides command to call frontend-slides for creation.

Frontend Slides can not only generate HTML presentation webpages directly, but also convert existing PowerPoint files into web-based presentations.

For example:

/frontend-slides I want to create a pitch deck for research

This can directly generate an HTML presentation webpage and open it in the browser.

/frontend-slides Convert my presentation.pptx to a web slideshow

This can convert a PPT file into an HTML presentation webpage.

This tutorial follows the first method and uses the Tricontinental dossier, The Turbulence of the Indian Economy, as a case study. During the generation process, Agent will provide several presentation generation options, which you can choose according to your needs.

Once generation is complete, the browser will automatically open the corresponding presentation. You can also continue giving instructions in the chat box to ask Agent to further modify the content.

AI-generated slide created by Frontend Slides, based on Tricontinental’s dossier The Turbulence of the Indian Economy

AI-generated slide created by Frontend Slides, based on Tricontinental’s dossier The Turbulence of the Indian Economy

Filling In a Pre-Designed Template

Another approach is to start from a pre-designed template and let AI fill in your content. This shifts the question from ‘how to generate content’ to ‘how to make the content look consistent and professional.’

In research presentations, one of the most common problems is not insufficient content, but rather a lack of visual consistency. Pre-designed template libraries solve this by providing a set of layouts that have already been visually unified. You can also create your own templates or design guides and save them for long-term reuse, building a consistent visual identity across your work over time.

One such library is beautiful-html-templates, which offers a large collection of HTML templates suitable for technical and academic scenarios, including data visualisation pages, dashboard-style layouts, research infographics, and methodology flowcharts. Each template has a clear hierarchy and unified structure, providing a level of consistency that default PowerPoint themes lack.

To use it, simply copy the following into Agent and run it:

Clone https://github.com/zarazhangrui/beautiful-html-templates and follow the instructions in AGENTS.md to build me a beautiful HTML slide deck.

As its name suggests, beautiful-html-templates is a project that contains many HTML templates. Open https://github.com/zarazhangrui/beautiful-html-templates to view reference examples of these templates.

The usage is also very straightforward. You only need to specify the template you want to use in Agent. For example:

Please use the Soft Editorial template to build a HTML slide deck for https://dev.thetricontinental.org/dossier-india-deindustrialisation/.

After Agent finishes executing, you can see that the presentation has been successfully generated.

AI-generated slide created by beautiful-html-templates, based on Tricontinental’s dossier The Turbulence of the Indian Economy.

AI-generated slide created by beautiful-html-templates, based on Tricontinental’s dossier The Turbulence of the Indian Economy.

PowerPoint Generation Tools

Automating Deck Generation in PowerPoint Format

Many academic deliverables still require PowerPoint as the output format — conference presentations, courses, thesis defenses, and workshops being among the most common. Tools that automate PowerPoint generation provide an alternative path: AI handles the content while the tool ensures the final output is a usable .pptx file.

This kind of tool is especially suitable for research teams and teaching scenarios. Trainers and teachers, for instance, can structure course materials and then batch-produce handouts or presentations with a consistent style, reducing repetitive layout work across multiple decks.

One such tool is PPT Master. To install it in Agent, enter:

Please help me configure this tool:

git clone https://github.com/hugohe3/ppt-master.git
cd ppt-master
pip install -r requirements.txt

After installation is complete, open the directory and you will see that the project has been successfully configured locally.

Then, enter:

Please turn the following into a PPT: [your document to convert]

And the previously generated HTML document can be converted into a PPT. The author also recommends placing the file to be converted in the ppt-master/projects directory for easier management.

Once the process is complete, the newly generated PPT file can be found in the exports folder.

AI-generated slide created by PPT Master, based on Tricontinental’s dossier The Turbulence of the Indian Economy.

General AI Design Systems

Generating Through a Design-System Skill

A more comprehensive approach is to use a full design system rather than individual templates. Where templates provide pre-set layouts, a design system provides a coordinated set of visual rules — alignment, margins, color schemes, typography hierarchy, and reusable components — that apply consistently across every page.

This is especially useful in research contexts, where content is often structurally repetitive. Methodology overviews, case studies, comparative analyses, and data visualisations share recurring patterns; once these patterns are encoded as design-system rules, AI can handle content generation while the system handles visual consistency.

One such skill is Huashu Design. To install it in Agent, enter:

Please help me install this skill:

npx skills add alchaincyf/huashu-design

After installation, you may need to restart Agent once to refresh the skill. You can then use the /huashu-design command to call huashu-design for creation.

For example, you can enter:

/huashu-design Make a HTML deck for https://dev.thetricontinental.org/dossier-india-deindustrialisation/.

Once generation is complete, you can see the final HTML presentation.

AI-generated slide created by Huashu Design, based on Tricontinental’s dossier The Turbulence of the Indian Economy.

AI-generated slide created by Huashu Design, based on Tricontinental’s dossier The Turbulence of the Indian Economy.

Using a Standalone Design Application

A standalone design application takes this further. Rather than living inside Agent as a skill, a dedicated design application provides a complete visual environment — component systems, visual asset management, collaborative editing, and unified design standards that can extend across slides, figures, posters, and other publication outputs. This direction is closer to an open-source version of Figma or Canva, with a stronger emphasis on openness and extensibility.

One such application is Open Design. Download the version for your operating system from: https://github.com/nexu-io/open-design/releases/tag/open-design-v0.8.0

After installation and configuration are complete, you can create a new Open Design project.

Next, enter the project name and choose the corresponding design system style.

Once the project is created, you can directly describe the presentation you want to generate in natural language.

After generation, an HTML chart preview will be displayed in real time on the right side.

AI-generated slide created by Open Design, based on Tricontinental’s dossier The Turbulence of the Indian Economy.

AI-generated slide created by Open Design, based on Tricontinental’s dossier The Turbulence of the Indian Economy.

Compared with traditional screenshot-based plotting, the biggest advantage of this approach is that the generated charts are truly editable, scalable, and reusable front-end charts. In the end, you can also export the result as PDF, PPTX, or HTML for use in papers, web pages, or presentations.

How to Edit Generated HTML

Although the generated HTML presentation usually already meets basic needs, in practical use it often still requires multiple rounds of adjustment. Fortunately, for AI, editing HTML has become very easy. Users do not need to master complicated syntax; they only need to describe their requirements accurately.

For example, if you want to redesign the layout of a specific page in the previously generated HTML document, you only need to enter:

Please help me redesign the layout of Page 03 in the file @india-deindustrialisation.html

After Agent finishes the task, you can return to the browser and see that the HTML presentation has been successfully updated.

Redesigned slide with an updated layout, based on Tricontinental’s dossier The Turbulence of the Indian Economy.

How to Find More Design Skills

AI is developing very quickly, and it is possible that before long, the tools introduced in this article will gradually be replaced by newer solutions. Therefore, at the end of this article, one more tool is introduced for automatically discovering new skills.

Open Agent and enter:

Please help me install this skill globally:

npx skills add https://github.com/vercel-labs/skills --skill find-skills

After Agent completes the installation, the system will successfully add a tool for searching skills.

This skill is highly practical because it can automatically find suitable skills based on the task requirements. For example, if you enter the following in the input box:

/find-skills Please help me identify the skills needed to create beautiful PowerPoint presentations and HTML slide decks.

Agent will automatically search for a large number of related skills, along with their download counts, brief descriptions, and installation methods.

If you are satisfied with the search results, you can also directly choose a corresponding skill to install. For example:

Please help me install and configure the frontend-slides skill.

This will automatically install the frontend-slides skill.

After Agent finishes executing, restart VSCode to refresh, and then you can start using the skill.

The Future of Research Presentations Will Increasingly Resemble Content Engineering

Taken as a whole, these open-source projects are pointing in the same direction: research presentations are gradually moving away from traditional manual production and toward a form of content engineering that is automatable, structured, and reusable.

The workflow of the future will probably no longer be ‘open PowerPoint and lay out each page one by one,’ but rather: first let AI understand the research content, then let open-source design systems generate slides automatically, and finally export them as PPT, PDF, or web-based presentations.

This means that what researchers need to focus on in the future may no longer be how to spend a large amount of time adjusting visual details, but how to express the logic of their research more clearly.

AI Is Redefining Scientific Visualization

For a long time, creating scientific figures has been an exhausting and time-consuming process. Many researchers have experienced the same workflow: repeatedly adjusting arrows in PowerPoint just to finish a flowchart, endlessly tweaking fonts and colors in Illustrator to maintain consistency, or reorganizing entire layouts simply to make a figure more readable. The real difficulty has never been ‘not knowing how to design,’ but rather the fact that scientific figures demand accurate information, clear structure, consistent styling, and professional visual presentation at the same time — and existing tools require enormous manual effort to get every detail right.

As a result, figure production has long been one of the most disproportionately costly steps in research communication: the time and energy it consumes often far exceed what the work itself should require. This is especially true for social science research in and on the Global South, where mapping commodity chains, capital flows, networks of power, or histories of struggle depends on visual communication that text alone cannot convey — yet access to professional design infrastructure has historically been out of reach. The emergence of AI has fundamentally changed this situation for the first time. Modern image-generation models are now capable of understanding module hierarchies, page layouts, workflow structures, scientific illustration styles, and even complex chart compositions. Even the long-standing issue of garbled text has improved significantly. This means researchers can finally spend less time manually arranging layouts and more time focusing on communicating ideas.

Scientific Figures Are Not All the Same

When many people first try AI image generation, they simply type something like ‘help me draw a scientific figure.’ In reality, however, scientific figures come in many different forms, and each type requires a completely different visual structure. The examples in this section are AI-generated figures based on Tricontinental’s dossier War on the Poor: A Frontline Report on the Effects of the Pandemic in the Global South, illustrating how each figure type can communicate different aspects of movement research.

Scientific Infographics

Scientific infographics function more like visual summaries. Their goal is not to present every technical detail exhaustively, but to help readers quickly grasp the research context, the problem being addressed, the core argument, and the key findings within seconds. Because of this, infographics often combine icons, modular layouts, concise text, color segmentation, and highlighted regions to establish clear visual hierarchy.

AI-generated infographic based on Tricontinental’s dossier War on the Poor.

Compared with traditional paper figures, infographics place much greater emphasis on readability and viewing experience. They are particularly suitable for graphical abstracts, dossier opening pages, briefing summaries, public-facing research communication, and overview pages for publications. AI performs especially well in this category because it is highly effective at constructing large-scale visual structures quickly. In many cases, researchers do not even need to think about colors at the beginning; they simply need to decide what story the figure should tell.

Workflow and Process Diagrams

Another extremely common category is workflow and process diagrams. Their defining feature is explicit procedural logic. For example, how a policy moves through institutions, how a crisis cascades across sectors, how capital circulates between actors, or how a historical process unfolds across stages — all of these are naturally suited for flow-based visualization.

AI-generated process diagram based on Tricontinental’s dossier War on the Poor.

In many situations, readers are not primarily interested in technical detail, but rather in understanding how a process actually unfolds. The greatest value of workflow diagrams lies in making complicated processes immediately understandable. Compared with infographics, they place stronger emphasis on sequence, input-output relationships, dependencies between stages, directional arrows, and structural hierarchy. Therefore, the key criterion is not whether the figure ‘looks beautiful,’ but whether the reading path is intuitive. A well-designed workflow diagram naturally guides readers from left to right or top to bottom without visual confusion.

Data Visualization

If workflow diagrams explain methods, data visualization explains evidence. In research publications, most conclusions must eventually be supported through figures: how inequality has shifted over time, how an indicator compares across regions, how a trend has evolved across a decade, or whether differences between groups are significant. Explaining these purely through text quickly becomes difficult to read.

AI-generated data visualization based on Tricontinental’s dossier War on the Poor.

At its core, data visualization transforms abstract numerical relationships into intuitive visual relationships. Truly effective charts do more than merely display data; they emphasize trends, amplify differences, control information density, and establish visual focus. The same data can appear either confusing or immediately understandable depending on how the figure is designed. This is one reason why serious research publications invest considerable effort into refining visual presentation, even when the underlying analysis is relatively straightforward.

An AI Workflow for Scientific Figure Generation

AI figure generation is an iterative process, not a one-shot result. AI is excellent at rapidly generating layouts and first drafts, but figures that meet publication standards still require human review and refinement.

Before writing any prompt, the workflow has three preparation steps. First, identify the figure type your work calls for — infographic, workflow diagram, or data visualization. Each organizes information differently: infographics emphasize conceptual integration, workflow diagrams emphasize procedural order, and data visualizations emphasize trends and differences. Second, define the structure — decide on the hierarchy: what the title, primary modules, annotations, and supporting elements are, and how they relate. Third, define the style — specify the color system, typography, emphasis regions, and reading order. The closer a prompt resembles a formal design specification, the more stable and reliable the generated result becomes.

Generating Through a Prompt-Assistance Skill

One approach is to use a prompt-assistance skill — a lightweight tool that reads your content and optimizes prompts for figure generation. As one example, the open-source baoyu-skills package provides a collection of such skills. To try it, open Agent and run:

Please help me install this skill:

npx skills add jimliu/baoyu-skills

After installation, restart VS Code and type /baoyu- to access a large collection of figure-generation skills.

You can also explore different visual styles at: https://github.com/JimLiu/baoyu-skills

Choose the figure-generation skill that best matches your preferred style and workflow.

Generating Through Configured Model APIs

A more integrated approach is to configure model APIs directly, so that your Agent can generate figures from prompts without going through a web interface. By setting up OpenAI or Google API keys locally, figure generation becomes fully automated — entering a prompt produces a completed figure on your machine. Current image-generation models such as GPT Image 2 and Nano Banana 2 handle structure and layout well enough to draft usable scientific figures from a single prompt.

OpenAI API keys can be obtained from: https://platform.openai.com/api-keys

Google image-generation API keys are available at: https://aistudio.google.com/app/api-keys

After obtaining the API keys, you can configure them inside Agent using the following command (replace [your api key] with your actual key):

Please help me configure [OPENAI_API_KEY/GEMINI_API_KEY]: [your api key]

Of course, if you only want a lightweight experience, using APIs directly may feel overly technical. If you want both Agent’s cross-file analytical capabilities and the convenience of web-based image generation, you can additionally include the following instruction inside Agent:

Please generate a detailed image-generation prompt for recreating the image.

This extracts the structured prompt generated by Agent, allowing you to paste it directly into a web-based image-generation interface. While this approach may not be fully automated, it provides a low-barrier way to experience the entire workflow from analysis to generation.

Using AI to Generate the First Draft

Although baoyu-skills can help restructure prompts, the original prompt itself still benefits from careful refinement. In practice, the most effective prompts do not simply pile together requirements. Instead, they resemble professional design specifications with clearly defined constraints. For example:

You are a scientific figure designer. Please generate a scientific workflow diagram of type [infographic/workflow/data visualization] with the topic '[your topic]'.

Requirements:

Clear information segmentation and intuitive reading order (top-to-bottom or left-to-right).
Titles should be prominent but not oversized; labels must remain readable without becoming too dense.
All text must be correctly spelled with no garbled characters or AI-generated mistakes; use sans-serif fonts for English and Heiti-style fonts for Chinese.
High image quality with clean details and no obvious editing artifacts or blurred regions.
Element proportions should follow realistic visual logic; avoid oversized arrows or tiny unreadable icons.
Maintain a unified academic publication style. Do not use cartoon or hand-drawn aesthetics. Ensure balanced alignment and composition.

The key point is not to let AI ‘freely improvise,’ but to clearly specify that the desired result is a publication-quality scientific figure rather than a generic illustration.

Open Agent, invoke baoyu-skills (for example, baoyu-infographic), and use the prompt above to generate an infographic. To make the generation process more transparent, append the following sentence at the end of the prompt:

Please generate a detailed image-generation prompt for recreating the image.

After Agent finishes execution, it will produce a highly detailed image-generation prompt specifying the figure type, layout, color system, content organization, and design rules.

Create a **Scientific Infographic** (visual summary / graphical abstract) following these specifications:

## Figure Type Definition
This is NOT a workflow diagram and NOT a data chart panel. It is a **Scientific Infographic**: a visual summary that lets the reader grasp within **seconds** — (1) research background, (2) what problem the analysis addresses, (3) core thesis/contribution, (4) headline result. Use **modular tiles, icons, concise text, color zones, and visual hierarchy** — not sequential process arrows, not detailed statistics charts.

## Image Specifications

- **Type**: Scientific Infographic (Graphical Abstract / Visual Summary)
- **Layout**: `bento-grid` — modular grid with varied cell sizes, hero tile + supporting tiles
- **Style**: Clean academic infographic — flat vector icons, color-coded modules, minimal text per cell. NOT cartoon, NOT hand-drawn, NOT workflow arrows, NOT chart-heavy
- **Aspect Ratio**: 16:9 (landscape — standard for graphical abstracts and poster headers)
- **Language**: English (sans-serif throughout; Heiti-like sans-serif for any Chinese)

## What This Figure Must Do

| Goal | How to achieve |
|------|----------------|
| Instant comprehension | Reader understands the dossier's argument in 5–10 seconds |
| Visual hierarchy | Hero tile largest; 5–6 supporting tiles with icons |
| Scannability | Max 1 headline + 2 short bullet lines per tile; no paragraphs |
| Emphasis | One hero statistic tile; color zones separate themes |
| What to AVOID | Process arrows, step numbering, bar/line charts, swimlanes, dense data tables |

## Core Principles

- **Modular, not sequential**: tiles can be read in any order; no mandatory arrow path between all cells
- **Icon-led**: each tile has one clear flat icon representing its theme
- **Text minimalism**: headline words + short phrases only; preserve key facts verbatim where used
- **Color zones**: each tile group uses distinct Bandung Circuits background tint to signal theme
- Ample whitespace between tiles; strict grid alignment
- NO cartoon, hand-drawn, watercolor, or illustrative narrative styles

## Text Requirements

- **CRITICAL**: Zero typos, garbled characters, or AI misspellings
- Title prominent but NOT oversized — balanced against tile grid
- Body per tile: max 2 lines, clearly legible, not dense
- **All English**: clean sans-serif (Helvetica, Arial, Inter, Poppins)
- **Chinese (if any)**: Heiti-like sans-serif (Source Han Sans)
- NO serif fonts

## Bandung Circuits Color System

| Role | Hex | Usage in this infographic |
|------|-----|---------------------------|
| Deep Indigo | `#1A0A2E` | Title, tile headlines, icon strokes — first color registered |
| Terracotta | `#C4572A` | Accent rules, tile borders, pull-quote bar |
| Golden Ochre | `#D4A03C` | Hero statistic numeral — sparingly |
| Sage Green | `#5A7A6B` | Campesino-impact tile background only |
| Warm White | `#FAF5EF` | Canvas |
| Parchment | `#F2E8D8` | Alternating tile fills |
| Charcoal | `#2A2A2A` | Body text |
| Muted Teal | `#2A7B88` | Policy-solution tile accent |
| Signal Red | `#D94F30` | "Myth vs Reality" contrast label — sparingly |

**Rules**: Max 3 colors per tile; **SAFFRON SAFEGUARD** — never orange-spectrum + Sage Green in same tile; no gradients, shadows, or gloss.

---
## Bento Grid Layout (6 Tiles + Header)

### Header Band (full width, slim)

- **Title**: "The War on the Poor"
- **Subtitle**: "Narcotics, Campesinos, and Capitalism"
- **Source**: "Tricontinental: Institute for Social Research"
- Deep Indigo title on Warm White; Terracotta thin rule below
---

### TILE A — Hero (2×1, top-left, largest)

**Icon**: Scales of justice tilted toward a bank building (flat line icon)
**Headline**: "Core Thesis"
**Text** (concise):
- "War on Drugs ≠ moral crusade"
- "Illicit money liquefies the banking system"
- "Imperial tool against peasants & sovereign states"
**Fill**: Parchment; Terracotta left accent bar
---

### TILE B — Background / Problem (1×1, top-right)

**Icon**: Globe with crossed-out "moral crusade" label
**Headline**: "The Problem"
**Text**:
- "Dominant narrative: drug trade is separate from capitalism"
- "Reality: underground circuits serve formal finance"
**Fill**: Warm White; Deep Indigo border
---

### TILE C — Research Lens (1×1, mid-left)

**Icon**: Simplified Colombia map outline + coca leaf
**Headline**: "Research Lens"
**Text**:
- "Colombia's coca-cocaine economy"
- "Campesinos at bottom of commodity chain"
**Fill**: Parchment

---

### TILE D — Hero Statistic (1×1, mid-center, visual emphasis)

**Icon**: Upward arrow with dollar sign
**Headline**: "Key Finding"
**Hero numeral** (Golden Ochre, large): **"62,000–80,000×"**
**Text**:
- "Price increase: farm gate → wholesale"
- "$0.56/kg → $44,580/kg"
**Fill**: Deep Indigo background; Warm White text; Golden Ochre numeral

---

### TILE E — Structural Insight (1×1, mid-right)

**Icon**: Two overlapping circles (legal / illegal) with dashed boundary
**Headline**: "Political Frontier"
**Text**:
- "Legal ↔ illegal drugs: not chemical, but political"
- "States protect corporate profits; criminalise peasants"
**Fill**: Warm White; Muted Teal icon accent

---

### TILE F — Human Cost (1×1, bottom-left)

**Icon**: Farmer figure with land (dignified, not pity framing)
**Headline**: "Who Pays the Price"
**Text** (Sage Green tint background — no terracotta here):
- "81% land held by top 1%"
- "57% poverty in coca-growing households"
- "Symptom, not cause — blocked subsistence"
**Fill**: Sage Green tint (#5A7A6B at 15% opacity on Parchment)

---

### TILE G — Policy Takeaway (2×1, bottom-right)

**Icon**: Seedling / land reform symbol
**Headline**: "Path Forward"
**Two-column mini contrast** (no arrows):
- ✗ Militarisation | Eradication | Criminalisation
- ✓ Land reform | Guaranteed prices | Confront laundering
**Fill**: Parchment; Muted Teal checkmarks; Signal Red crosses

---

### Footer (full width, slim)

"Source: Tricontinental — The War on the Poor (CC BY-NC 4.0) | thetricontinental.org"
Dark Gray, sans-serif regular
---

## Text Labels (verbatim where data appears)

**Title block**: "The War on the Poor" | "Narcotics, Campesinos, and Capitalism" | "Tricontinental: Institute for Social Research"
**Tile headlines**: "Core Thesis" | "The Problem" | "Research Lens" | "Key Finding" | "Political Frontier" | "Who Pays the Price" | "Path Forward"
**Key data**: "62,000–80,000×" | "$0.56/kg → $44,580/kg" | "81% land top 1%" | "57% poverty"
**Footer**: "Source: Tricontinental — The War on the Poor (CC BY-NC 4.0)"

---

## Final Quality Checklist

- [ ] Reads as **visual summary**, NOT workflow, NOT data dashboard
- [ ] 6 modular bento tiles + header + footer; no sequential step arrows between all tiles
- [ ] Each tile: icon + headline + max 2 short text lines
- [ ] Hero statistic "62,000–80,000×" visually dominant in Tile D
- [ ] Bandung Circuits colors; Saffron Safeguard observed
- [ ] Sans-serif only; zero typos; academic publication quality
- [ ] No cartoon/hand-drawn; clean alignment and balance

You can then continue using AI to generate a complete scientific infographic draft. For example:

AI-generated infographic produced from the prompt above, based on Tricontinental’s dossier War on the Poor.

Human Review Still Matters

Even though AI can now generate surprisingly mature figures, human review remains essential. The first draft usually needs careful inspection: text accuracy, font hierarchy, color consistency, whitespace, arrow thickness, border weight, and shadow usage all require attention. In many cases, what determines whether a figure truly ‘looks like a paper figure’ is not the core content itself, but these subtle details.

The good news is that the amount of manual work required is now dramatically smaller than drawing everything from scratch. You can use Illustrator, Photoshop, or even PowerPoint to perform final refinements on exported figures. In other words, AI provides speed, while humans provide judgment. This is currently the most practical collaborative workflow.

HTML Is Becoming a New Approach to Scientific Visualization

For complex data visualization, pure image generation is not always the best solution. Many scientific charts require precise control over axes, data relationships, layout proportions, and export quality. In these situations, asking AI to generate HTML or JavaScript chart code is often more stable, more controllable, and easier to modify later.

One particularly recommended tool is Open Design. It is essentially an AI-driven visualization platform capable of generating HTML charts directly from natural-language descriptions while providing real-time previews. Its greatest advantage is that the output is not a fixed image, but an editable, scalable, and exportable visualization page.

Download the appropriate version for your operating system from: https://github.com/nexu-io/open-design/releases/tag/open-design-v0.8.0

After installation and configuration, create a new Open Design project.

Then enter the project name and select the desired design system style.

Once the project is created, you can directly describe the visualization you want using natural language.

For example:

Please generate a high-quality scientific data visualization chart using HTML. The overall style should resemble figures from Nature or IEEE papers.

Topic:
[your topic]

Data:
[your data]

Requirements:
White background, strict alignment, unified typography, no cartoon style, natural spacing between bars, all values clearly readable, and no animations or shadows.

After generation, the HTML chart preview will appear in real time on the right side.

Compared with traditional screenshot-based workflows, the greatest advantage of this approach is that the charts are genuinely editable, scalable, and reusable frontend visualizations. Finally, you can export the results as PDF, PPTX, or HTML for use in papers, websites, or presentations.

Conclusion

AI is dramatically lowering the barrier to scientific figure creation. In the past, producing professional scientific graphics required substantial design experience. Today, researchers can quickly generate visually coherent and structurally clear scientific illustrations. More importantly, however, the real transformation is not simply that ‘drawing figures has become faster,’ but that scientific communication itself is changing.

For the theory and methodology behind what makes an infographic work, see the companion article Demystifying Infographic Design.

Future research workflows will likely evolve into a collaborative system where AI handles understanding and generation, design systems maintain stylistic consistency, and researchers focus on scientific accuracy and aesthetic judgment. Scientific figures are no longer just supplementary illustrations inside papers; they are gradually becoming central components of research communication itself. This may be the perfect time to try using AI to create the first truly professional figure for your next paper.

AI can now rapidly generate large volumes of ‘paper analyses,’ but much of this content is merely fluent text with an academic tone. They look like critical studies, yet they lack clear sources of evidence, analytical pathways, and verifiable bases for judgement.

Genuine critical evaluation is not impromptu commentary or emotional negation, but a structured process. It requires the researcher to clarify the analytical objectives, fix the evaluation criteria, and ensure that every judgement can be traced back to the source text. The focus of this article is not on making AI ‘better at critiquing,’ but on how to establish a set of critical analysis methods that are traceable, verifiable, and reusable over the long term.

Why AI Tends to Produce ‘Pseudo-Critique’

AI’s biggest problem is not that it cannot analyse, but that it is all too easy for it to ‘look like it is analysing.’ Without structural constraints, the model will automatically complete the logic, amplify vague cues, and generate a large number of standardised judgements such as ‘the research has limitations,’ and ‘the argument is insufficiently developed.’

But the real question is: do these conclusions actually come from the source text? Can they be located in specific paragraphs, evidence, or argumentative structures? If they cannot be verified against the original text, then the so-called ‘critical analysis’ is often merely speculation cloaked in an academic tone, rather than a genuinely reliable research evaluation.

How to Use AI for Critical Evaluation

Truly valuable academic critique is never an impromptu expression, but a rigorous analytical pipeline. It requires the researcher, before initiating any analysis, to first clarify the analyst’s identity and boundary of responsibility, then fix the analytical objectives, analytical stance, and output format, and then proceed step by step along a traceable path, so that every judgement can eventually be returned to the source text.

1. Role Definition

Many people habitually enter a vague instruction into the model, such as ‘please critically analyse this paper.’ But the consequence is that the model does not know whether it should become a summariser, a commentator, or a genuine critical analyst.

What the role setting determines is not the language style, but how the model understands the boundaries of its responsibilities. A model clearly defined as a ‘critical analyst’ will proactively seek out vulnerabilities in the argument, logical gaps, and evidential weaknesses. Therefore, the starting point of a critical evaluation must be to make the model clearly aware of who it is, what it is to do, and what professional standards it must follow. For example:

## Your Role

You are an expert **critical analyst** tasked with conducting a rigorous, constructive critical analysis of argumentative texts.

Your analysis must be faithful to the source text, focus on logical structure, evidence, and argumentative resilience, and avoid injecting unsupported opinions. 

The ultimate goal is to help strengthen the work by identifying areas for improvement.

2. Objectives and Audience

After defining the role, it is necessary to establish the specific objectives and context for the analysis work. If the requirement is only a vague ‘critique,’ the model is highly likely to slide between abstract, evaluation, and templated limitations, producing content that cannot be verified. Therefore, it is essential to lock in advance the purpose of the analysis, the identity of the intended reader, and the tone of the entire report.

The goal of the analysis is not simply to point out errors, but to conduct a ‘stress test;’ the audience may be the author, editor, or peer reviewer; the tone should be scholarly, precise, and fundamentally constructive. Once these dimensions are fixed, the entire subsequent analysis process has a unified direction.

## Objectives and Output Goals

### Primary Objective
Describe the purpose of the critical analysis, such as performing a stress test on the text's arguments, identifying weaknesses, and suggesting actionable improvements.

### Target Audience
Describe who the analysis is intended for (e.g., author, editor, peer reviewer) and what level of detail and rigour is expected.

### Required Tone
Describe the analytical tone: critically rigorous yet constructive, scholarly, precise, and clearly anchored in the source text.

3. Using a Reference & Navigation Section

In long prompts, models may lose track of where key instructions and source materials are located. They may skip intermediate sections, introduce external assumptions, or fail to follow the intended analytical scope.

Therefore, the prompt should begin with a clear navigation declaration that explicitly tells the model where the analytical rules and source text are located

This section functions as a structural “table of contents” and routing guide for the model, helping maintain instruction adherence across long-context processing.

## Methodological Reference and Input

This prompt is self-contained for its critical analysis task. 

All analytical criteria are provided within the **Detailed Critical Analysis Protocol** section.

Load the primary text to be analysed from the **Input** section at the end of this prompt. 

Do not consult any external references.

4. Execution Protocol

Next, the execution process of the critical analysis needs to be broken down into a set of rigorous requirements, forming the core operating system of the entire prompt. This part begins with a clear task statement, establishing an action framework for the model so that it can identify, examine, and measure each argument through repeatable steps.

The analysis protocol can usually contain three levels:

First, require the model to locate the core arguments and vulnerable claims from the text.

Second, stipulate for each argument the assessment lenses that must be applied, such as logical structure, evidential foundation, rhetorical clarity, and argumentative resilience.

Finally, clarify the analytical mindset: maintain a ruthlessly critical stance while always aiming to be constructive, and apply the ‘most damaging plausible interpretation’ to ambiguous statements to test argumentative strength. You can customise the types and depth of the assessment lenses according to your own research needs.

## Detailed Critical Analysis Protocol

### Argument Identification

Identify up to [number] central arguments and any particularly vulnerable claims in the provided text.

### Assessment Criteria

For each argument, systematically examine it through the following lenses:

- **Logical Structure:** Identify fallacies, assess validity and soundness of reasoning.
- **Evidentiary Foundation:** Evaluate the quality, relevance, and sufficiency of supporting evidence.
- **Rhetorical Clarity:** Detect ambiguity, equivocation, or manipulative language.
- **Argumentative Resilience:** Check for unaddressed counterarguments and internal consistency.

### Analytical Approach

Maintain a ruthlessly critical yet constructive stance. 
Apply the 'most damaging plausible interpretation' to ambiguous claims in order to test robustness.

For each analysed argument, provide: a concise summary, a detailed critical analysis linked to the criteria above, a vulnerability rating (1–10), and actionable suggestions for improvement.

5. Output Format

The rigour of the analysis process must ultimately be solidified through the structure of the output. If the model only generates a prose-style commentary, tracing the basis for any particular judgement becomes extremely difficult. Therefore, the prompt needs to specify a clear output format.

For example, require the output as a single Markdown document, providing for each analysed argument a summary, critical analysis, vulnerability rating, and improvement suggestions, and conclude with a comprehensive executive summary. At the same time, the language and academic style to be used can be stipulated here.

## Output Format

1. Produce the critical analysis as a **single, well-structured Markdown document**.
2. Begin with the title of the input document.
3. For each analysed argument, include: Argument Summary, Critical Analysis, Vulnerability Rating (1–10), and Improvement Path.
4. Conclude with an executive summary of the most critical overarching weaknesses.
5. All output must be presented in clear, precise academic English.

6. Input Content

After everything is ready, the final step is to paste the complete paper, chapter, or notes into the input section at the end of the prompt. It is this action that activates all the preceding silent rules, making the text the sole factual source for every judgement the model makes. Therefore, the text you provide must be complete; do not omit any paragraphs, footnotes, or data that might conceal key arguments.

## Input

`[Paste the FULL TEXT of the essay, academic article, or argumentative work to be critically analysed here.]`

Worked example

The sections above are not only a tutorial outline. they are a prompt template you can turn into a single file. When you are done, you have a full critical-analysis prompt, one self-contained instruction set the model can follow every time.

Once you have settled on that prompt, you can register it as a skill in your toolchain so you do not paste the whole protocol on every run. In VS Code with the Claude Code extension, for instance, you can ask the agent to install the file as a project skill and then invoke it on a document in one message, like the following:

Please help me install @project/How to Analyse and Critique Research with AI/critique-research.md as a skill in this project, and then use this skill to Analyse and Critique the article @project/How to Analyse and Critique Research with AI/test.pdf.

Here the model returned a layered memo, argument blocks, ratings, and an executive summary, rather than a single slab of prose.

Against a vague one-liner such as ‘Please help me Analyse and Critique xxx,’ the structured prompt keeps the output easier to scan and easier to map back to sections of the paper you are reviewing.

Visual Presentation

Pure Markdown text has several limitations: its formatting is relatively uniform, important points are difficult to emphasize visually, and the final appearance often depends heavily on the editor or rendering environment.

To improve readability and presentation quality, you can use the following prompt to transform Markdown into a richly styled standalone HTML document. This allows the content to be displayed in a cleaner, more structured, and more visually engaging way. For example, you can use the prompt below to convert your own content.

You are an expert editorial web designer, front-end developer, and long-form publication formatter.

Your task is to convert the provided markdown document into a polished, standalone HTML reading experience suitable for a premium think-tank, academic journal, or strategic analysis publication.

Requirements:

- Single self-contained HTML file
- Embedded CSS + JavaScript only
- Responsive layout
- Elegant academic/editorial aesthetic
- Serif body text, sans-serif headings
- Sticky left-side table of contents
- Reading progress bar
- Smooth scrolling
- Collapsible major sections
- Active TOC highlighting

Structure:

- Title area with metadata
- “Helicopter View” overview section
- Main numbered sections with concise subsection structure
- Minimal but clear visual hierarchy
- Subtle callout boxes for key critiques
- Simple vulnerability/risk badges
- Footer note

Style:

- Calm, intellectual, modern
- Long-form editorial publication aesthetic
- Avoid dashboard UI or generic markdown rendering
- Use restrained editorial colors and generous spacing

Content rules:

- Preserve the original structure and analytical tone
- Do not heavily rewrite content
- Do not over-summarize
- Reduce excessive bullet points where possible
- Prefer concise analytical prose with selective emphasis

Goal:
Transform the markdown into a publication-grade analytical article rather than a raw markdown export.

After submitting the prompt, wait for the Agent to process the content. Once generation is complete, you will be able to preview the rendered HTML result directly.

Conclusion

High-quality critical research is, in essence, a structured process of evidence analysis. It is not simply ‘finding fault,’ but testing a paper’s logic, evidence, and argumentative boundaries through systematic stress testing.

In this process, the most suitable role for AI is not to replace the researcher in making judgements, but to help organise the structure, identify problems, and solidify the analytical workflow. Only when all judgements can be traced back to the source text and re-verified does AI critique truly hold research value.

Appendix

AI can generate a text that ‘looks like a summary’ in seconds, but the real problem is not whether the model can write, but that many outputs are not suitable for research, teaching, or long-term knowledge management. They are often linguistically fluent and structurally complete, yet lack faithful preservation of the original text’s logic, context, and evidential provenance.

A truly usable literature summary does not merely compress content into something shorter, but preserves the original argumentative structure and conceptual relations as far as possible, and makes key conclusions traceable back to the source for verification. This article therefore focuses not on writing ‘longer’ prompts, but on how to build a repeatable, checkable, and maintainable process for AI-assisted summarisation — specifically of long-form academic texts such as journal articles, book chapters, monographs, and lecture or interview transcripts. The same scaffolding does not transfer cleanly to news, marketing copy, or short-form web content; those genres call for different prompts.

Why Many AI Summaries ‘Look Right but Are Actually Unusable’

The problem with many AI summaries lies not in obvious errors, but in ‘vague accuracy.’ The model will automatically complete the logic, smooth out arguments, and rewrite statements carrying limiting conditions into seemingly definitive conclusions. This problem is exacerbated when the model summarises based only on abstracts, partial segments, or a mixture of several sources.

Another core issue is the loss of structural alignment with the source. Many summaries silently collapse the original’s chapter and section organisation into a single tidy stream of prose — readers cannot tell which part of the summary corresponds to which part of the source, and any later attempt to revisit a specific argument is forced to re-read the original from scratch. They are suitable for rapid browsing but cannot truly enter serious knowledge-production workflows.

How to Use AI to Summarise Articles

The walk-through below builds the prompt in five numbered sections. Some sections are fixed — their wording carries the design decisions that make this prompt work, and rewriting them tends to break the prompt rather than tailor it. Others are variable — placeholders that only become useful when you fill them with content for your own scenario. Before reading the sections themselves, it helps to know which is which:

• 1 Role — Mostly fixed. Keep the ‘chronological, narrative-focused summary’ framing and the ‘self-contained yet faithful’ wording — these are what set the deliverable type. The list of genres (article, book chapter, podcast, talk transcript) can be trimmed or extended to match what you actually summarise.
• 2 Objectives and Audience — Variable. You fill in the Primary Objective, Target Audience, and Tone. The example shown in that section is a worked sample from the source template — read it as a benchmark of specificity, not as text to copy verbatim.
• 3 Execution Protocol (3.1–3.8) — Fixed. Keep the eight sub-protocols as written; they encode the load-bearing rules of the design. The one parameter commonly tuned is the length cap in §3.1 (10,000 words is the source template’s default — lower it for short articles, raise it for monographs).
• 4 Output Format — Fixed. Keep as-is unless your downstream pipeline requires a different shape (e.g., JSON instead of Markdown).
• 5 Input — Variable. Replace the placeholder with your actual source text on each run.

1. Role Definition

When people first try to get AI to analyse a paper, they often simply input a line such as ‘please summarise this article.’ Such blunt instruction may appear efficient, but in fact it pushes the model into a vague decision space.

For AI, ‘summarising’ in itself is not a task with clear boundaries: it could become a superficial listing of keywords, a compression of information lacking clear hierarchy, and more often than not, it devolves into vacuous correctness. The root of the problem is that we have not delineated clear boundaries of responsibility for the AI at the very beginning of the interaction.

A model that has been explicitly assigned the identity of an ‘expert summary assistant,’ with duties and limitations defined, will produce output of markedly different depth and structure compared to a generic model without role setting, even when given the same text.

It is worth being specific about which kind of summary this role is for. In this guide we focus on a chronological, narrative-focused summary — one that follows the author’s own ordering of ideas rather than reorganising them into a critique, a structural blueprint, or a cross-text comparison. Those other goals deserve their own dedicated prompts, and the role definition for each will differ accordingly. Naming this constraint at the role level prevents the model from drifting into analysis or restructuring later on.

Therefore, the starting point for any high-quality summary is to clothe the model in a professional identity and let it work with a clear self-understanding. For example:

## Your Role

You are an expert **summary assistant** tasked with producing a **chronological, narrative-focused summary** of an academic article, book chapter, podcast interview, or talk transcript. 
Your primary goal is to capture the text's **flow and content** from beginning to end in a way that is **self-contained** yet entirely **faithful** to the original.

You should avoid hallucinating information or injecting your own opinions. 
You must strictly follow all instructions regarding scope, structure, and faithfulness.

2. Objectives and Audience

After defining the role, it is necessary to establish more specific objectives and context for the task. The intended users and ultimate purpose of the summary determine its form. Therefore, this section should clearly fill in the primary objective of the summary, the target audience, language style, and so on.

A common failure here is to leave the audience description vague — ‘researchers and students,’ for instance. The more concrete the reader profile, the better the model can calibrate vocabulary, density, and what it assumes the reader already knows. As a benchmark for the level of specificity to aim at, the source template behind this guide names its audience as advanced doctoral students in fields such as philosophy, historical sociology, Marxism, intellectual history, and anthropology — a single line that tells the model both the disciplinary register and the assumed level of background. Your own audience line should be roughly that specific.

A worked example, lifted directly from the source template, looks like the following. Treat it as a reference shape rather than something to copy verbatim — keep the three-block structure (Primary Objective / Target Audience / Required Tone), but rewrite the bullets to match your own scenario.

## Objectives and Output Goals

### Primary Objective of the Summary

- Provide a **sequential, comprehensive** overview of the source text's arguments, ideas, and evidence as presented by the author.
- Serve as the reader's single best tool for understanding how the text's **main points evolve** and connect from start to finish.
- Minimise interpretive detours: **focus strictly on summarising** the original text rather than analysing, critiquing, reorganising, or adding external information.

---

### Target Audience for the Summary

- Advanced doctoral students in fields such as **philosophy, historical sociology, Marxism, intellectual history, and anthropology**.
- Readers who require a **straightforward, meticulously structured, and accurate** grasp of the text's argumentation **in the precise order it is presented by the author**.

---

### Required Tone of the Summary

- Academic, clear, precise, and coherent.
- Neutral and objective, **without** extensive theoretical framing or rhetorical analysis of the source.
- A thorough and faithful retelling that preserves the source text's natural progression and emphasis.

3. Execution Protocol

Next, the execution protocol for the summarisation needs to be broken down into a series of rigorous sub-protocols, forming the core operating system of the entire prompt. This is the longest and most consequential part of the prompt: rather than collapsing it into a single block of ‘do these things,’ it should be structured as a numbered protocol the model executes step by step. Each sub-protocol corresponds to a concrete decision the model would otherwise make implicitly — and inconsistently.

The eight sub-protocols below mirror the lifecycle of producing a summary, from setting scope, through structuring and writing, to final self-checks.

3.1 Scope and Length

Specify which parts of the original must be covered, and how long the summary can be. Two things matter. First, scope: cover all major sections, arguments, and significant references in the order they appear, and do not compress in ways that would distort the author’s emphasis or the overall structure. Second, length: give an upper bound (e.g. up to 10,000 words for a long, dense source) but ask for conciseness where it does not sacrifice completeness.

3.2 Required Structure

This is where most generic summaries collapse — they read fluent but feel structureless. Force the model to follow a fixed scaffold:

• Helicopter view: open with one or two short paragraphs stating what the text is about, how it is organised (a brief roadmap of its sections), and any key themes that will recur.
• Sequential walk-through: organise the body strictly in the same order as the source. Mirror the original’s section headings as your sub-headings, and use bullets, short paragraphs, or numbered subsections to break down each section’s arguments, evidence, examples, and transitions.
• Key elements to surface throughout: major claims and theories; definitions of concepts and terms as the author uses them; significant references (thinkers, texts, events the author invokes); author-provided context or background; examples or case studies used to illustrate arguments.
• Conclusion: briefly restate the overall thesis as the author articulates it, plus any closing thoughts, implications, or calls to action.

3.3 Quotations and Paraphrasing

Direct quotes are valuable but expensive. Permit short direct quotes only for emblematic, defining, or striking passages that uniquely illuminate the author’s style, a key assertion, or a critical definition. Prefer precise paraphrase for general arguments. If the source text is not in English, state explicitly whether the summary is quoting an existing translation or producing its own.

3.4 Clarity and Explanatory Notes

Two rules in tension. On one hand, break dense or abstract arguments into clear, accurate explanations suitable for the intended audience, and define specialist terminology as the author uses it. On the other hand, do not import external critiques, your own analysis, or theoretical frameworks not present in the original. The only exception is minimal background context: if the author leans on an obscure reference (a thinker, event, or prior debate) without explaining it, a brief parenthetical clarification is allowed — but only to make the author’s own argument intelligible, never to introduce new analysis.

3.5 Faithfulness and Completeness

Three commitments. No omissions or additions: do not skip major sections or supporting points, and do not insert interpretations not found in the source. Order and logic: rigorously follow the sequence in which the author presents ideas — and if the original is itself disjointed or non-linear, reflect that faithfully rather than silently tidying it up. Accuracy: dates, names, events, and direct quotes must match the source exactly.

3.6 Style and Formatting

Specify the academic register and how the summary should be navigable. Use Markdown headings and subheadings liberally to mirror the source’s structural divisions. Use transitional phrases that reflect the author’s stated logic — for example, ‘Building on this point, the author then discusses…’ or ‘Having established X, the text turns to Y…’ — so that the reader sees how the original moves, not just what it says.

3.7 Deliverable Requirements

State what the finished artefact should be: a self-contained summary that lets a reader unfamiliar with the original gain a cohesive, start-to-finish understanding; a chronological, narrative flow that reports each portion in the order originally presented. Equally important is stating what this prompt is not — it is a narrative summary, not an ideological critique, a rhetorical dissection, or a comparative analysis. Those are separate prompts in the same toolchain (the companion TextKillerCritique template in this directory is one such); keep this one strictly to the linear ‘story of the text’ approach. This boundary is what lets multiple prompts coexist without overlapping.

There is also a quieter boundary worth making explicit: the deliverable is a standalone narrative document, not a traceback or research-aid index. Structural alignment with the source comes from mirroring section headings (see §3.2 and §3.6), not from per-paragraph page numbers or [Section X]-style anchors attached to each claim. If your downstream use genuinely needs that kind of precise pointer back to the original — for citation harvesting, argument-by-argument verification, or evidence auditing — treat it as a separate task with its own prompt, rather than stretching this one to cover it.

3.8 Final Check

Close the protocol with an explicit self-check the model must run before delivering:

• Adherence to the specific task: no drift into ideological positioning, rhetorical analysis, or comparative tables.
• Complete yet streamlined: all essential content and arguments covered, without interpretive commentary.
• Ready for graduate-level reference: polished enough to serve as a reliable, stand-alone resource for the intended audience.

Putting it together, the protocol section of the prompt looks like the following skeleton:

## Detailed Summarisation Protocol

You are to execute the summarisation according to the following detailed protocol.

### 1. Scope and Length

Specify which parts of the source must be covered and the maximum length of the summary, balancing completeness against conciseness.

---

### 2. Required Structure

**Helicopter View**

Open with a short paragraph stating what the text is about, how it is organised, and the key themes that recur.

**Sequential Walk-Through**

Organise the body in the source's own order, with sub-headings mirroring the original's structural divisions.

**Key Elements to Surface Throughout**

Major claims, definitions of concepts and terms, significant references, author-provided context, and any examples or case studies.

**Conclusion**

Restate the author's overall thesis and any closing thoughts or implications.

---

### 3. Quotations and Paraphrasing

Reserve direct quotes for emblematic or defining passages; prefer precise paraphrase elsewhere. 
Specify how translations are to be handled.

---

### 4. Clarity and Explanatory Notes

Break down dense arguments and define terms as the author uses them. Do not import external critiques. 
Allow minimal parenthetical context only for unexplained obscure references essential to the author's argument.

---

### 5. Faithfulness and Completeness

No omissions or additions. Follow the source's sequence and logic — including any disjointed or non-linear structure. 
Ensure factual accuracy of dates, names, events, and quotes.

---

### 6. Style and Formatting

Academic register; Markdown headings mirroring the source; transitional phrases that reflect the author's own logical moves between sections.

---

### 7. Deliverable Requirements

A self-contained, chronological, narrative summary. 
Not an ideological critique, not a rhetorical dissection, not a comparative analysis — those belong to separate prompts in the toolchain.

---

### 8. Final Check

Before delivery, verify: no drift into other task types; complete yet streamlined; polished as a stand-alone reference for the intended audience.

4. Output Format

The prompt also needs to set clear output format requirements, for example that the output file must be in Markdown format, strictly follow the text-formatting conventions specified in the prompt, and clearly state that no external citations or bibliography are required. This serves as the systematic packaging of all the preceding specifications, ensuring the model knows how to organise them into a deliverable whole.

## Output Format

1. Produce the summary as a **single, well-structured Markdown document**.
2. Adhere strictly to the output specification, especially regarding the use of headings and subheadings to mirror the source text.
3. No external citations or bibliography are required. The focus is on the input text. Output as pure markdown text.

5. Input Content

Finally, in the input section at the end of the prompt, the complete paper, chapter, or transcript needs to be pasted in as the primary content, activating the entire process.

## Input

`[Paste the FULL TEXT of the academic article, book chapter, podcast interview transcript, or talk transcript HERE for summarisation. 
Ensure the text is clean and readable.]`

Worked example

The sections above are not only a tutorial outline; they are a prompt template you can turn into a single file. When you are done, you have a full summarisation prompt, one self-contained instruction set the model can follow every time.

Once you have settled on that prompt, you can register it as a skill in your toolchain so you do not paste the whole protocol on every run. In VS Code with the Claude Code extension, for instance, you can ask the agent to install the file as a project skill and then invoke it on a document in one message, like the following.

Please help me install @project/How to Summarise Papers and Books with AI/summarise-papers.md as a skill in this project, and then use this skill to summarise the article @project/How to Summarise Papers and Books with AI/test.md

The model then produced a long-form summary with clear headings and a stable thread through the source’s own order of ideas.

By contrast, a minimal line such as Please help me summarise xxx tends to return something that reads fluent but thin on structure and weak on tying claims back to the text. The same document, run with the explicit prompt above, came back with sectioning you can skim and content that hangs together against the original.

Validating the Output

Section 3.8 of the protocol is the model’s self-check. It catches certain drift patterns — sliding into critique, padding with interpretation, going off-format — but it cannot tell you whether the summary is actually a faithful representation of the source. The model has no way to compare itself against the original. That comparison is the reader’s job, and it deserves a short, repeatable routine of its own. Without it, the whole pipeline is open-loop: you have a process that should produce a good summary, but no mechanism that confirms a given run did.

The failures worth checking for are rarely ‘the summary is wrong in obvious ways.’ They are the quiet ones:

• Section coverage. Open the source’s table of contents alongside the summary. Every major section of the source should be locatable in the summary as a sub-heading or a clearly demarcated block. Missing sections are common when the source is long, when its chapter structure is weak, or when one section is far denser than the others.
• Hedge preservation. Pick three to five claims in the summary that read as confident assertions. Find the corresponding passages in the source. Models have a well-documented tendency to silently rewrite hedged statements (‘the author tentatively suggests…’, ‘X may be the case under conditions Y…’) as flat declaratives. If the source had qualifiers and the summary has dropped them, that is over-confidence drift — exactly the ‘vague accuracy’ failure mode the introduction warned about.
• Definition fidelity. For any specialist term the author introduces or uses in a stipulative sense, check that the summary’s gloss matches the author’s usage rather than the term’s general meaning. Over-smoothing of stipulative definitions is the single most common quality failure on theoretically loaded texts.
• Quote accuracy. Direct quotes should match the source word-for-word, including punctuation. Random spot-check two or three.
• Phantom references. Sample one or two of the ‘significant references’ the summary names — thinkers, texts, events. Confirm the author actually invoked them, in roughly the role the summary describes. Hallucinated references are rare in modern models but not zero, and they are catastrophic if they survive into downstream citation.

Budget roughly ten to fifteen minutes for this the first few times. After a handful of runs on the same kind of source you will learn which failure mode your specific genre is prone to, and the routine compresses to whichever subset still earns its time.

Visual Presentation

Pure Markdown text has several limitations: its formatting is relatively uniform, important points are difficult to emphasize visually, and the final appearance often depends heavily on the editor or rendering environment.

To improve readability and presentation quality, you can use the following prompt to transform Markdown into a richly styled standalone HTML document. This allows the content to be displayed in a cleaner, more structured, and more visually engaging way. For example, you can use the prompt below to convert your own content.

You are an expert editorial web designer, front-end developer, and long-form publication formatter.

Your task is to convert the provided markdown document into a polished, standalone HTML reading experience suitable for a premium think-tank, academic journal, or strategic analysis publication.

Requirements:

- Single self-contained HTML file
- Embedded CSS + JavaScript only
- Responsive layout
- Elegant academic/editorial aesthetic
- Serif body text, sans-serif headings
- Sticky left-side table of contents
- Reading progress bar
- Smooth scrolling
- Collapsible major sections
- Active TOC highlighting

Structure:

- Title area with metadata
- “Helicopter View” overview section
- Main numbered sections with concise subsection structure
- Minimal but clear visual hierarchy
- Subtle callout boxes for key critiques
- Simple vulnerability/risk badges
- Footer note

Style:

- Calm, intellectual, modern
- Long-form editorial publication aesthetic
- Avoid dashboard UI or generic markdown rendering
- Use restrained editorial colors and generous spacing

Content rules:

- Preserve the original structure and analytical tone
- Do not heavily rewrite content
- Do not over-summarize
- Reduce excessive bullet points where possible
- Prefer concise analytical prose with selective emphasis

Goal:  
Transform the markdown into a publication-grade analytical article rather than a raw markdown export.

After submitting the prompt, wait for the Agent to process the content. Once generation is complete, you will be able to preview the rendered HTML result directly.

Conclusion

The greatest value of AI is not to replace reading, but to help humans organise reading more efficiently. A high-quality academic summary is, in essence, a sequence of disciplined moves rather than a single generative act: first set the scope (what kind of summary, for whom, in what register), then fix the structure (the eight sub-protocols that govern coverage, order, faithfulness, and self-check), then generate, and finally validate against the source rather than against your own sense that the output ‘reads right.’

When AI is placed inside that four-beat rhythm — scope, structure, generate, validate — it stops being a text machine that returns whatever sounds plausible, and starts being a piece of research infrastructure: outputs you can trust enough to cite, hand to a collaborator, or build a literature note on top of.

Appendix

The sample input, the prompt, and the generated output (in both Markdown and HTML) are available in the Bandung Circuits repository open-sourced by Global South Insights.

Why Scanned PDFs Are a Problem

Plenty of academic papers, historical documents and official records exist only as scanned images, page after page of locked‑up pixels that standard text‑based tools can’t touch.

When your research involves pulling quotes, tables or references from dozens of such files, manual transcription becomes a serious bottleneck. Optical Character Recognition (OCR) engines exist, but stitching them into a reliable, repeatable AI workflow is far from trivial: most raw OCR outputs lose headings, merge columns or scramble table structures, which makes it impossible for a large language model to reason over the document with any confidence.

You need a tool that not only ‘reads’ the image but also reconstructs the document’s logical layout and hands back clean, structured, machine‑friendly text. That is exactly what Docling is built to do.

What Docling Does

Docling is an open‑source document understanding library that converts a wide array of document formats into structured information AI agents can consume. Far more than a simple OCR wrapper, it processes materials where text, tables and images are interleaved, including multi‑column layouts, hand‑written notes and complex PDFs that are really just scanned photographs.

Key capabilities that set Docling apart:

• Mixed‑content understanding, it handles documents where natural reading order does not match the raw PDF stream. Tables stay tables, figures retain their captions, and headers are not dumped into the body.
• Format flexibility, the library accepts PDF, DOCX, PPT, XLSX, HTML, images (PNG, JPEG, TIFF), LaTeX, plain text, WAV, MP3, and WebVTT. This dramatically expands the range of source material an AI agent can work with downstream.
• MCP integration, through the Model Context Protocol (MCP), Docling becomes a service that your agent can call directly. This closes the ‘upload → convert → query’ loop within a single conversation, without you ever leaving the chat interface.

In short, Docling takes a photograph of a page and turns it into a Markdown file that preserves the true structure of the content. That structure is what makes the difference between an AI that hallucinates a plausible‑sounding summary and one that actually cites the right paragraph.

Installing Docling Locally

The steps below use an agent in VS Code, but the same approach works in any environment that supports MCP.

First, open your agent’s chat panel.

Agent panel

Modern AI agents can often locate and apply installation instructions on their own. To get started, enter a prompt that describes exactly what you want:

Please help me install the Docling MCP service for use with an AI agent
and configure the environment if necessary. 
For configuration, refer to the documentation at https://docling-project.github.io/docling/usage/mcp/

MCP status shows Connected after manual steps

Once the agent confirms that the configuration is complete, Docling is ready to use.

Agent confirms installation is complete

You can verify the installation by typing /mcp in the dialog. This lists all MCP servers currently connected to your agent.

Type /mcp to see the service list

When docling appears in the MCP list and its status is Connected (usually shown in green), the service is active and waiting for your requests.

Docling appears as Connected in the MCP list

To test the setup, upload a sample scanned PDF and ask the agent:

Please convert this PDF to Markdown.

Prompt to convert a PDF to Markdown

A successful conversion will return clean, structured Markdown, headings, paragraphs, tables and image references all preserved. This confirms that your agent can reliably read and process the PDF’s content, which is the foundation for later tasks such as summarisation, key‑point extraction, and evidence‑grounded question answering.

A successful Markdown conversion result

If the conversion fails or the output looks garbled, ask the agent to check the logs and suggest corrections, common fixes include adjusting the system dependencies or re‑scanning the original PDF at a higher resolution.

Checking Output Quality

Obtaining a Markdown file is only half the job. For research‑grade work, you must verify that the output faithfully represents the original document. Spend a few minutes examining a representative sample and pay close attention to the following points:

• Paragraph integrity and reading order, check that key paragraphs are complete and appear in the same logical sequence as in the PDF. Multi‑column layouts sometimes confuse OCR engines, so look for text fragments that may have been shuffled.
• Table fidelity, confirm that tables are identified and that their row‑column structure is preserved. A table rendered as a flat string of numbers is almost useless for downstream analysis.
• Mathematical formulas and special characters, symbols, Greek letters, and block equations are error‑prone. Verify that they have been converted accurately and are not mangled or omitted.
• Headers and footers, ensure that running heads, page numbers, and footnotes are not incorrectly merged into the main text, which would pollute the extracted content.

For particularly important documents, it is safest to cross‑check the Markdown output against the original PDF page by page. If you spot systematic issues, such as unrecognised fonts or consistently broken tables, try re‑scanning the document at a higher DPI or choosing a different export format before sending it to Docling. Small adjustments to the input often yield dramatically better results.

Summary

Docling transforms locked‑up scanned content into structured, machine‑readable Markdown, bridging the gap between static documents and AI‑grounded research. With out‑of‑the‑box support for PDF, DOCX, PPT, XLSX, HTML, PNG, JPEG, TIFF, LaTeX, plain text, WAV, MP3, and WebVTT, it removes the format chaos that typically bogs down document‑heavy projects. Once the MCP service is installed and verified, you can simply describe any conversion or analysis task to your agent and let it handle the rest.

When combined with retrieval‑augmented generation (RAG) or knowledge graph construction, the clean text that Docling produces becomes reliable fuel for evidence‑based answers. Instead of guessing what a scanned paper might say, your AI has the actual text, and that makes all the difference between plausible noise and trustworthy insight.

For simple projects, collecting sources by hand and writing prompts is usually enough. When you need a full research pass that spans dozens or hundreds of papers, the limits of manual search and download become clear.

This article frames grounded research around two complementary moves: web searching (programmatically discovering what exists on the open web and in scholarly indexes) and web fetching (retrieving full documents or machine-readable representations through documented interfaces you control). Used with each provider’s terms of service and rate limits, that combination speeds collection and lowers hallucination risk by tying model outputs to sources you can verify.

Web Searching Versus Web Fetching

Traditional search engines return links and short snippets. Humans can follow those links one by one; large-scale or agent-driven workflows usually need structured search results, fields you can parse, filter, and hand to the next step.

Here, web searching means calling search-oriented APIs so the agent gets normalized data (titles, URLs, snippets, ranks, and similar) instead of rendered HTML.

The JSON below is a simple illustration of what structured search output can look like:

[
{
"title": "Google",
"link": "https://google.com/",
"snippet": "Search the world's information, including webpages, images, videos and more. _Google_ has many special features to help you find exactly what you're looking ...",
"position": 1
},
{
"title": "Google Scholar",
"link": "https://scholar.google.com",
"snippet": "_Google Scholar_ provides a simple way to broadly search for scholarly literature. Search across a wide variety of disciplines and sources: articles, theses, ...",
"position": 2
}
]

Web fetching is the next step: obtaining full text or cleaned page content, a PDF, or a Markdown-like extraction of a page you are allowed to process, so the model can work from complete arguments, not previews alone.

Together, structured search and disciplined fetching form a pipeline from query to evidence, the practical backbone of grounded research.

Why These Capabilities Matter

Generative models hallucinate when context is thin or wrong. Feeding them retrieved, checkable material, metadata, abstracts, PDFs, or excerpts from permitted channels, anchors answers in real sources.

For end-to-end research projects, programmatic search and fetch help because they deliver:

1. Traceability: URLs, identifiers, or API payloads you can log and revisit.
2. Richer context: Full text or clean extractions outperform snippets alone when nuance matters.
3. Freshness: Many academic and web APIs surface newer records than a model’s static training cutoff.

What follows surveys tools you can connect with API keys and published docs, plus a few interactive aids where the value is exploration rather than automation.

Web Searching

The sections below cover how to discover candidates, papers, pages, datasets, before you commit to full retrieval.

1. Serper.dev

Serper.dev belongs to the same broad family as other SERP-style APIs: you send an HTTP request with a query and credentials, and you receive structured JSON instead of clicking through a results page. Engines, parameters, and quotas are defined in Serper’s documentation; the day-to-day workflow is familiar if you have already used SerpApi-style products.

Start by creating an account: https://serper.dev/signup.

After you verify your email, the site typically drops you into the playground at https://serper.dev/playground. There you can inspect recent usage and try queries interactively.

To call the API from code or from an agent, open the API Keys tab in the left sidebar. A key is usually generated for you automatically; copy it and store it securely (for example in an environment variable).

For Google Search via Serper, follow the HTTP method, headers, and body format in the official docs when you call the API from code or from an agent. As a quick sanity check, you can sometimes open a minimal URL in the browser (replace {query words} and {your api key}):

https://google.serper.dev/search?q={query words}&apiKey={your api key}

The response is JSON, not a consumer-style HTML results page, exactly what programmatic clients expect.

In your agent, keep the key in something like SERP_DEV_API_KEY and reference it from prompts instead of hard-coding it. For example:

Please use the SERP_DEV_API_KEY from the environment variables to search for information about 'Google' with the Google search engine and provide a summary. 
The API format is:
https://google.serper.dev/search?q={query words}&apiKey={your api key}

The agent can then call the endpoint and summarize the JSON.

Serper’s free tier is generous on volume (on the order of thousands of searches for new accounts at the time of writing), but it exposes fewer engines and surfaces than SerpApi. Choose Serper when a narrow, Google-centric stack is enough; reach for SerpApi when you need additional Google products (Scholar, News, and others) from one vendor.

2. SerpApi

SerpApi wraps Google and many other Google ‘engines’ as structured data, so agents can automate discovery without opening each result by hand. It turns result pages into clean JSON that is easier to filter, rank, and pass downstream than raw HTML.

Register and sign in at https://serpapi.com/users/sign_in.

The default free plan includes a limited number of searches per month; upgrade if your workload grows.

From the dashboard, copy your API key.

A minimal Google Web request looks like this (substitute your query and key):

https://serpapi.com/search?engine=google&q={query words}&api_key={your api key}

Paste the URL into a browser once to confirm you see JSON rather than HTML.

In the agent, read the key from the environment, for example SERP_API_KEY, and keep the URL pattern in the prompt or in tool configuration:

Please use the SERP_API_KEY from the environment variables to search for information about 'Google' with the Google search engine and provide a summary. 

The API format is:
https://serpapi.com/search?engine=google&q={query words}&api_key={your api key}

The model issues the call and returns a short summary of the hits.

That pattern generalizes: SerpApi also exposes Google Scholar, Google Events, Google Finance, and more, adjust engine and the prompt so the agent picks the right surface.

Web Fetching

Searching tells you what might be worth reading. Fetching is how you obtain full content in compliant ways, PDFs from open repositories, metadata from scholarly APIs, or normalized text from tools built for extraction, not improvised bulk copying of entire sites.

PDF and document-oriented APIs

1. arXiv

arXiv is an open-access preprint service operated by Cornell University. It is widely used in computer science, physics, mathematics, and neighboring fields. Because authors often upload before journal publication, arXiv is frequently where new work appears first.

arXiv supports both discovery and retrieval: you can search metadata and obtain PDFs through stable, documented HTTP interfaces. The public metadata API is described in the arXiv help center; the API basics / quickstart explains that clients issue ordinary HTTP requests and receive XML feeds whose entries include links to PDFs.

You do not need to parse the XML by hand in every workflow, your agent or a small script can, but it helps to know that the response is structured, not a single human-facing HTML page.

Example query (metadata for up to ten results whose metadata matches electron in the default ‘all’ fields):

http://export.arxiv.org/api/query?search_query=all:electron&start=0&max_results=10

The returned feed lists titles, summaries, and related links; PDFs are reachable from the entry links as described in the official documentation. Respect arXiv’s published terms and request etiquette.

In an agent, you might keep the interaction high-level:

Please search arXiv for papers related to electrons and download the PDFs locally.

The agent should implement the XML request, pick entries, and download only what the user or policy allows.

2. Semantic Scholar

Semantic Scholar, from the Allen Institute for AI, pairs semantic ranking with rich metadata (authors, venues, citations, related work). Its APIs support both literature search and, where licensing and publisher links allow, retrieval of PDFs or landing URLs, always subject to each paper’s access rules.

Sign in at https://www.semanticscholar.org/sign-in.

Use the website search to sanity-check titles, authors, or topics.

Programmatic access requires a separate API key. Apply at https://www.semanticscholar.org/product/api.

After approval, you receive a key by email.

Store it as something like S2_API_KEY and point the agent at the official client examples:

Please use the S2_API_KEY stored in your environment variables to search for articles about 'xxx' using the Semantic Scholar API and download the PDFs locally where permitted. 
For API usage, refer to the documentation at https://github.com/allenai/s2-folks/tree/main.

Web page content

Some teams chain search to hosted extraction services that turn allowed URLs, or small, configured jobs, into clean text or Markdown for downstream models.

Firecrawl and Crawl4AI are commonly cited in agent tutorials: you supply URLs and constraints; the tool returns normalized content. They are not a substitute for reading each site’s robots policy, copyright notice, and terms of use. This article does not lay out crawling tactics; stay within what each vendor documents and what the target site permits.

Other tools: Connected Papers

Connected Papers is an interactive map: from one seed paper it draws a graph of related work so you can spot clusters and high-influence nodes before you fix a reading list. It is not mainly an API-first fetch layer, but it is excellent for seeing how papers connect.

Open https://www.connectedpapers.com/ and search by title.

The tool builds a visual graph of neighbors in citation and similarity space.

Conclusion

Web searching through Serper.dev and SerpApi gives you structured web discovery; Semantic Scholar does the same for scholarly metadata. Web fetching, arXiv’s XML API for open preprints, plus Semantic Scholar where keys and licenses allow, fills in full text and pointers. Connected Papers supports exploratory mapping of the literature, while tools such as Firecrawl and Crawl4AI can supply normalized page text only when policy and contracts allow.

How to Talk to AI – A Prompt Engineering Guide

Why You Need Prompt Engineering

Most dissatisfaction with AI assistants isn’t because the model isn’t ‘smart enough,’ but because there’s a disconnect between the task you have in mind and the signals the system can actually process. The same model might seem mediocre when you ask it a random question, but when you clearly define the task, its boundaries, and the expected output format, it often performs quite well.

For example, if you simply say, ‘Help me write an email,’ the system has to guess the recipient, tone, length, taboos, and what constitutes a ‘good’ email. This guesswork spreads the probability across many possible responses, resulting in answers that are often vague and off-target. This isn’t some mystical phenomenon – it’s the standard behavior of generative models when constraints are insufficient.

To avoid this divergence, the key is to specify the constraints clearly: who the audience is, what the purpose is, what the hard boundaries are, and what constitutes a successful delivery. You don’t need to write a long essay, but you must make it clear to the model ‘which answer is the right one.’

Common Markdown in Prompts

What is Markdown?

Markdown is a lightweight writing format. Unlike Microsoft Office, which relies on software-defined formatting, it uses simple symbols (such as #, -, and **) to structure a document.

For example:

• # Title → creates a heading
• – item → creates a list
• **text** → makes text bold

Why Use Markdown?

LLMs cannot interpret software-defined formatting; they only process raw text. However, raw text alone can be disorganized and difficult to reuse. Using simple symbols to structure your content makes it clearer and accessible to both you and the LLM.

You don’t need to switch your entire writing workflow to Markdown, but familiarizing yourself with a few of its features can help you write prompts that read like ‘mini-documents’ – easy to read, edit, and reuse.

Use Symbol to Structure Your Prompt

Headings break down the ‘role,’ ‘output rules,’ and ‘source material’ into sections, making it easier for both the model and humans to skim. When you paste the same prompt into an AI input box, the structure remains intact.

# Prompt guide

## role

You are a helpful assistant...

Bold text is used to highlight non-negotiable rules, field names, or acceptance criteria.

**Don't generate low-quality text**

Code blocks (three backticks) enclose examples, JSON, YAML, or ‘edit-only’ snippets to reduce the likelihood of them getting mixed up with instructions.

Please return the content in the following way: 

```
{
    ‘output’: ‘<response content>‘
}
```

Ordered or Unordered Lists are ideal for writing steps, acceptance criteria, or ‘must include / must not contain’ lists.

1. Content should be high quality.

2. Don't generate garbled characters or meaningless symbols.

- Content should be high quality.

- Don't generate garbled characters or meaningless symbols.

Tables are suitable for comparisons, scoring criteria, and listing pros and cons.

|     Correct      |     Incorrect     | 
| ---------------- | ----------------- | 
| This is correct  | This is incorrect |

By using these elements, you can turn a simple wish into a document-like prompt, rather than just a single line of text floating in a chat.

How to Render Markdown Files

As shown above, Markdown is easy for LLMs to process, but less convenient for humans to read in its raw form. To improve readability, you can use tools that render Markdown into a formatted view, such as Obsidian or Visual Studio Code.

For example, in Visual Studio Code, you can search for ‘Markdown’ in the Extensions marketplace and install a suitable extension to preview and render Markdown files. For details, please refer to Set Up Your AI Workspace.

What’s the Difference Between Casual Chit-Chat and Markdown Templates?

The syntax itself isn’t the goal; the difference lies in whether you’re willing to turn your tasks into reusable, iterable templates. Templates can’t replace thinking, but they can capture your thoughts so you don’t have to start from scratch every time.

The concept often discussed in education – the shift ‘from prompt engineering to knowledge engineering’ – has a similar core idea: If explanations, reference materials, and checklists are all preserved as structured text, they become more like team assets rather than just personal insights. Here’s an example that illustrates the difference between a weak prompt and a strong prompt.

Context, Actions, Outputs

In prompt engineering, a good prompt is often broken down into three parts. This follows the same logic as the ‘writing Markdown in sections’ mentioned earlier, just using a different notation:

• Context describes what material you’re working with or what situation you’re in (file names, paths, what the data roughly looks like).
• Action: Clearly state what needs to be done, avoiding vague, unverifiable phrases like ‘clean it up’ or ‘optimize it’.
• Output: Specify what the result should look like (filename, format, paragraph structure, required fields). Only when all three sections are filled out will the model not have to guess your underlying assumptions.

The most common issue is writing too vaguely. You can place these three sections under headings like ## Context / ## Action / ## Output to make skimming easier.

A More Structured Approach

A more reliable approach is to break down the process into distinct modules and write prompts based on their structure:

• Opinion & Insight: The top-level must be controlled by humans, responsible for objectives, core judgments, and accountability, rather than being entirely delegated to the model.

The five modules below collectively support this layer:

• Problem Orientation: Clearly define the problem to be solved, as well as the objectives and direction.
• Ideology: Outline the fundamental stance, writing principles, or content to be avoided.
• Information Reserve: Organizes background materials and available facts.
• Methodology: Defines the execution process and specific steps.
• Form of Expression: Defines the tone, structure, and formatting.

These modules are interdependent: without a clear problem orientation, even abundant information can easily stray from the topic; without a defined form of expression, even the best methodology may produce incoherent output.

You can use this framework to write each section in Markdown, helping large language models make more consistent judgments rather than relying on a single vague instruction. Below are some comparative examples:

Problem Orientation

Ideology

Information Reserve

Methodology

Form of Expression

From a ‘Single Prompt’ to a ‘Multi-Step Chain’

When a task is relatively simple, it is usually sufficient to include all the information in a single prompt.

However, when the task involves retrieval, filtering, multiple rounds of revisions, or strict requirements, a single prompt presents two problems:

• First, it becomes too long, making it difficult to read and increasing the risk that the model will overlook certain parts.
• Second, it makes it difficult to pinpoint errors; when the results deviate, it’s hard to determine which step went wrong.

A more reliable approach is to break the process down into multiple steps and save each one separately. For example: Step 1 defines the problem or outline, Step 2 organizes the information, and Step 3 generates the final content. The output of each step is a structured result, which serves as the input for the next step.

This ensures each step is clear and controllable, and facilitates individual review and modification.

Three-Step Research Example

Below is a simple three-step research example to illustrate the prompt chain:

Step 1: Please generate an analytical framework for researching this topic

# Framework Designer

## Your Role

You are a research framework designer. Your task is to design a structured analytical framework for the given topic.

## Input

**Research Topic**: {RESEARCH_TOPIC}

## Task

Design an analytical framework for the topic ‘{RESEARCH_TOPIC}’.

### Framework Design Principles

...

## Output Format

Write to file: `workspace/{topic_id}/01.framework/framework.md`

...

Step 2: Conduct research and summarize by section for each dimension of the framework

# Dimension Researcher

## Your Role

You are the Dimension Researcher. Your task is to conduct in-depth research and summarize a specific dimension within the analytical framework.

## Input

Read `workspace/{topic_id}/01.framework/framework.md` to obtain:
- **Dimension ID**: {DIMENSION_ID}
- **Dimension Name**: {DIMENSION_NAME}
- **Core Question**: {CORE_QUESTION}
- **Key Analysis Points**: {ANALYSIS_POINTS}
- **Research Topic**: {RESEARCH_TOPIC}
- **Output Path**: {OUTPUT_PATH}

## Task

Conduct web searches and gather materials for the assigned dimension, then write a research summary for that dimension.

### Research Process

...

## Output Format

Write to the file: `workspace/{topic_id}/02.findings/{dimension_id}.md`

...

Step 3: Merge all sections into a single structured report and save it with the specified filename

# Report Assembler

## Your Role

You are the Report Assembler. Your task is to merge all dimension research summaries into a single, structured, comprehensive report.

## Input

- **Framework file**: `workspace/{topic_id}/01.framework/framework.md`
- **Dimension summary directory**: `workspace/{topic_id}/02.findings/`
- **Output filename**: {OUTPUT_FILENAME}
- **Research topic**: {RESEARCH_TOPIC}

## Task

Read all dimension summaries and assemble them into a final report according to the framework structure.

### Assembly Process

...

## Output Format

Write to file: `workspace/{topic_id}/03.report/{OUTPUT_FILENAME}`

...

This is a typical three-step prompt chain with three types of deliverables, rather than a single, overly long prompt. The researcher can ultimately retrieve the complete report at

workspace/{topic_id}/03.report/{OUTPUT_FILENAME}

Let AI Write the Prompt

Once you’ve mastered structured writing, a more efficient approach is to first have the model generate a reusable Markdown prompt based on your general goal, and then standardize the wording, add facts, and refine the scope.

Here’s an example request you can copy and paste directly, with just a few minor adjustments:

You are a prompt engineer. I will describe my goal in rough natural language. First, assess whether the goal is clear; if it is ambiguous or incomplete, ask up to three clarifying questions. Then, produce a reusable Markdown prompt template that includes: role definition, context, the input I will provide, output format, quality checks, and stopping conditions.

In terms of the workflow, the model typically provides the basic structure of the template first, and you are then responsible for verifying:

Whether the facts are accurate, whether they comply with specifications, whether proper nouns and data are correct, and whether the template contains any unverified assumptions.

This way, you don’t need to waste energy on making the ‘opening perfect,’ but can instead focus on quality control and iteration. This division of labor is more akin to knowledge organization and verification: the model builds the framework, while the human is responsible for the final result.

An Example of an AI-Generated Prompt

For example, if I want to study prompt engineering, I can provide the AI with the following prompt:

You are a prompt engineer. I will describe my goal in rough natural language. First, assess whether the goal is clear; if it is ambiguous or incomplete, ask up to three clarifying questions. Then, produce a reusable Markdown prompt template that includes: role definition, context, the input I will provide, output format, quality checks, and stopping conditions. 

I want to conduct research on prompt engineering.

After answering the questions posed by the AI, you can generate the prompt template. As you can see, the AI helps us organize our thoughts and generates a clearer, reusable, and complete prompt.

## Role

You are a research scientist specializing in human-AI interaction and natural language processing, with expertise in decomposing complex reasoning tasks through structured prompt engineering techniques.

## Context

Prompt chaining is a technique where complex tasks are broken into sequential subtasks, with each step's output feeding into the next prompt. This approach improves reliability, interpretability, and performance on multi-step reasoning problems. Your research aims to synthesize best practices, evaluate chain architectures, and produce actionable guidance for researchers implementing these systems.

## Input I Will Provide

- Domain focus (e.g., scientific reasoning, code generation, multi-document analysis)
- Baseline task description or example of a complex problem to decompose
- Available models/tools (e.g., Claude, GPT-4, domain-specific APIs)
- Constraints (latency budgets, cost limits, interpretability requirements)
- Prior work (specific papers, frameworks, or failure cases to consider)

## Output Format

For each research query, produce:

1. **Decomposition Strategy** – Diagram or description of the proposed chain structure (nodes = prompts, edges = data flow)  
2. **Prompt Templates** – Draft prompts for each node with explicit input/output contracts  
3. **Evaluation Protocol** – Metrics for assessing chain performance vs. monolithic baseline  
4. **Failure Mode Analysis** – Anticipated error propagation points and mitigation strategies  
5. **Implementation Sketch** – Pseudocode or architectural notes for the chain execution logic

## Quality Checks

- Each chain node has a single, well-defined responsibility  
- Inter-node data contracts are explicit (typed inputs/outputs)  
- Error handling is defined for each transition  
- The chain is justified over alternatives (parallel calls, single-shot, or tool use)  
- Evaluation includes both end-to-end accuracy and per-step diagnostics  

## Stopping Conditions

Stop and request clarification when:
- The domain requires specialized knowledge beyond general NLP (medical, legal, safety-critical) without provided source materials  
- The proposed chain exceeds 10 nodes (suggests need for hierarchical refactoring or alternative architecture)  
- No evaluation data or success criteria are specified  

---

Summary

Collaborating with AI essentially involves clearly defining the task first, then using a simple structure to align expectations between both parties. Markdown isn’t just a format; it allows humans and models to share the same ‘instruction manual.’ If the model builds the framework first while you control the facts and boundaries, you can boost efficiency while avoiding a loss of control over the results.