research-capture
Skill5 starsChanged 34 days agoMITView on GitHub
Turns a link, article, paper, video transcript, book passage or chat conversation into a research note in Knowledge/ that keeps what the source says separate from the user's own view, records where every claim came from, and links related notes. Use when the user says "save this article", "take notes on this paper", "research X and write it up", "capture what we just figured out", "add this to my knowledge base", runs /research-capture, or pastes a URL with "remember this". Not for meeting notes (use meeting-notes) or decisions (use decision-record).
1 other project uses a version of this.
Adopted by
Most recently taken up by
- lukasrepublic/agentic-foundrytheir own version, first seen Oct 11, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- lukasrepublic/agentic-foundry★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- decision-recordSkill
- inbox-triageSkill
- log-feedbackSkill
- meeting-notesSkill
- onboardingSkill
- project-briefSkill
- weekly-reviewSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 20
Ask, or infer from context: what does the user want to know or remember?
- Distinctive: line 38
Before finishing, run the checks in references/source-checklist.md.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Research capture
A research note is useful later only if you can tell what a source said, what you concluded, and how much to trust each.
1. Get the material
- URL: fetch it. If it cannot be fetched (paywall, login, error), say so and ask the user to paste the text. Never write a note from the title or from memory as if you had read the source.
- File (PDF, transcript): read the parts that matter to the user's question.
- A topic to research: search, then read the sources you will cite. Prefer primary sources over summaries of them.
- A conversation: the source is the conversation; note that it is unverified.
2. Find the question
Ask, or infer from context: what does the user want to know or remember? A note organized around a question is easier to reuse than a summary of a whole source.
3. Check for an existing note
Grep Knowledge/ for the topic. If a note exists, add to it (new source, new claims, updated view) instead of creating a duplicate.
4. Write the note
From System/Templates/Research-Note.md, in Knowledge/Title-Case-Name.md:
- Question: one line.
- What the sources say: claims, each followed by its source. Quote only when wording matters, and mark quotes.
- What I think: the user's view if they gave one; otherwise leave a prompt for them, or offer your own reading clearly marked as yours.
- Open questions: what is still unknown, including anything you could not verify.
- Related: typed links (
related_to::,part_of::) to projects and other notes. - Frontmatter:
sources:lists every URL or reference;status: draftuntil the user has reviewed it.
Before finishing, run the checks in references/source-checklist.md.
5. Hand over
Give the user a three-line summary and the link to the note. Name any claim you could not verify and any source you could not open.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

