agentleFS
Sign inStart free
umararshad4umararshad4/honey-stack/skills/research/SKILL.md

research

Skill4 starsChanged 15 days agoMITView on GitHub

Investigate a question using primary sources and leave a cited, reusable research note.

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 165 edited versions we read keep these.

  • 3

    Distinctive

    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
  1. Distinctive:

    Read the user's constraints and relevant project docs.

    line 17
  2. Distinctive:

    Use available Exa tools for deep, current, multi-source or code discovery work.

    line 25
  3. Distinctive:

    Use the note format for investigations.

    line 35

“Common” is measured: kept by at least 6 in 10 of the 165 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Research

Answer one decision or question from sources the reader can inspect. For the actual APIs of already-declared dependencies, use opensrc; for wider questions, comparisons, unfamiliar technology or disputed facts, use this workflow. Research does not authorize implementation, account creation or publication.

1. Frame the decision

Read the user's constraints and relevant project docs. Write the question, the hard exclusions, the date-sensitive claims, and what evidence would settle the decision. An option that fails a hard constraint cannot win the ranking.

Choose the smallest useful output: a direct answer for a narrow lookup; a reusable note for a multi-source investigation. Use the repo's existing research location, otherwise docs/research/YYYY-MM-DD-<topic>.md. When repo writes are not authorized, keep the answer in chat or the host's scratch-artifact location.

Done when: the scope and exclusions are explicit, and the destination matches the request.

2. Find the sources

Use available Exa tools for deep, current, multi-source or code discovery work. If unavailable, say so once and use the available search tools. Follow discovery results to primary sources: official documentation, source repositories, specifications, first-party APIs, papers or original datasets. Use gh for GitHub reads.

Read the relevant pages or source files, not just snippets. Check installed versions and source revisions for technical claims, and both publication and event dates for changing facts. Record inaccessible sources; do not fill their contents from memory. Conflicting authoritative sources remain a disagreement until version, date or context resolves it.

Done when: each important claim has a source that directly supports it, including relevant version/date context.

3. Compare and qualify

Separate measured results, source statements and your inferences. Cross-check decisive claims with another authoritative source when one exists; if only one source supports a claim, label that limitation rather than inventing corroboration. A project's README and its marketing page are not independent confirmation.

Use the note format for investigations. Cite the supporting source beside each claim, include tradeoffs and rejected options, and explain what evidence would change the recommendation. Keep secrets, private session quotes and customer data out of external search queries and shared notes.

Done when: the recommendation satisfies the hard constraints, and uncertainty is visible beside the affected claim.

4. Hand off

Give the answer first and link the note if one was created. grill can use unresolved product choices; plan can consume confirmed constraints and findings. Re-check unstable facts before later implementation rather than treating the note as permanently current.

Done when: the reader can act on the answer and find its evidence without repeating the research.

Download now (ZIP)

  • Other versions187 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 187

    1. mattpocock/skills★ 309ksame name
    2. Yeachan-Heo/oh-my-claudecode★ 40ksame name
    3. Yeachan-Heo/oh-my-claude-sisyphus★ 40ksame name
    4. QuivrHQ/quivr★ 40ksame name
    5. udecode/plate★ 17ksame name
    6. openai/plugins★ 7.4ksame name
    7. ArcReel/ArcReel★ 6ksame name
    8. vinvcn/mattpocock-skills-zh-CN★ 5.5ksame name
    9. withcoral/coral★ 4.9ksame name
    10. docker/docs★ 4.7ksame name

    177 more not listed

    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.

  • Used alongsideProjects with it also use daily-arxiv, exp-eval and 10 more.

    Other files people use alongside this one.

    Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
    • LambenthanSkill

      daily-arxiv

      每日从 arXiv 拉取新论文，过滤相关性，auto-ingest 高优先级论文，检测 SOTA 更新

      Also in 5 projects that use this

    • LambenthanSkill

      exp-eval

      实验判决门：Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges

      Also in 5 projects that use this

    • truongtv22Skill

      Problem-Solving Techniques

      Apply systematic problem-solving techniques for complexity spirals (simplification cascades), innovation blocks (collision-zone thinking), recurring patterns (meta-pattern recognition), assumption constraints (inversion exercise), scale uncertainty (scale game), and dispatch when stuck. Techniques derived from Microsoft Amplifier project patterns adapted for immediate application.

      Also in 5 projects that use this

    • liufeng1976Skill

      setup-ts-deep-modules

      Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.

      Also in 33 projects that use this

    • parcadeiSkill

      ast-grep-find

      AST-based code search and refactoring via ast-grep MCP

      Also in 4 projects that use this

    • parcadeiSkill

      debug-hooks

      Systematic hook debugging workflow. Use when hooks aren't firing, producing wrong output, or behaving unexpectedly.

      Also in 4 projects that use this

    • +5 more from Lambenthan
  • In this repository39 other files it gives its agents.

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free