save-research-doc
Skill49 starsChanged 3 months agoMITView on GitHub
Save a piece of research, notes, or content to the user's filesystem as a well-organised markdown document with metadata. Use when the user asks to save, archive, file away, or "keep this for later".
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
After producing a substantial result the user wants to retain
- Distinctive: line 22
All lowercase, hyphens not spaces
- Very specific: line 45
Use offer_choices: "Open the file", "Save another version", "Add to a topic index", "Done"
- Distinctive: line 49
Always add the date — future you needs to know when this was true
- Distinctive: line 53
Don't save without permission if the conversation didn't explicitly ask
- Very specific: line 54
Don't overwrite without warning.
- Distinctive: line 55
Don't save sensitive data (passwords, keys, personal info that shouldn't be persisted)
- Distinctive: line 56
Don't dump raw chat transcript — synthesise the useful content
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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
Save Research Document
When to use
- "Save this for me"
- "File this under [topic]"
- "I want to keep this research"
- After producing a substantial result the user wants to retain
Steps
- Identify what to save — usually it's the most recent assistant output. If ambiguous, confirm via
ask_questions. - Decide on path and filename:
- Top-level folder by topic:
research/,notes/,briefs/,code/,drafts/ - Filename with date for time-sensitive content:
YYYY-MM-DD-slug.md - Otherwise descriptive slug:
analyse-pricing-strategy.md - All lowercase, hyphens not spaces
- Top-level folder by topic:
- Add frontmatter — the document gets YAML frontmatter for searchability:
--- title: [Document title] created: [ISO date] tags: [tag1, tag2] summary: [One-sentence summary] sources: [optional list of URLs cited] --- - Format the body:
- H1 title at the top
- Source content cleaned up (preserve markdown formatting)
- Inline source links if research-based
- Footer with "Last updated: [date]"
- Write the file — call
fs_writewith the path and content. - Confirm and offer follow-ups:
- Show the path and size via plain message
- Use
offer_choices: "Open the file", "Save another version", "Add to a topic index", "Done"
Style
- Generous structure. Use headings, bullet points, dividers — make it scannable later.
- Always add the date — future you needs to know when this was true
- Cross-link related saved docs if you know about them (call
fs_listto check)
What not to do
- Don't save without permission if the conversation didn't explicitly ask
- Don't overwrite without warning. Check
fs_readfirst; if exists, ask viaconfirm_action - Don't save sensitive data (passwords, keys, personal info that shouldn't be persisted)
- Don't dump raw chat transcript — synthesise the useful content
In this repository32 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- agent-asks-tasksSkill
- caretakerSkill
- code-reviewSkill
- compare-optionsSkill
- create-reportSkill
- csv-analyseSkill
- document-qaSkill
- draft-emailSkill
- enrich-errorSkill
- extract-structured-dataSkill
- fact-checkSkill
- git-diff-summariserSkill
- librarian-curateSkill
- meeting-notesSkill
- morning-briefSkill
- plan-taskSkill
- reflectSkill
- remember-conversationSkill
- reverieSkill
- review-code-securitySkill
- review-email-toneSkill
- review-outputSkill
- review-summary-faithfulnessSkill
- rewrite-for-audienceSkill
- route-findingSkill
- routine-health-checkSkill
- score-importanceSkill
- summarise-urlSkill
- web-researchSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.

