agentleFS
mementomori-comementomori-co/maker-mastery/.claude/skills/research-topic/SKILL.md

research-topic

Skill2 starsChanged 25 days agoMITView on GitHub

Use when you want a topic researched into verified notes. Splits the topic into narrow questions, fans out researcher agents in parallel, then runs /verify on each inbox file and /file on each verified file; lead-run, it does not write notes itself.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read CURRICULUM.md and the knowledge\ category indexes to pick the category the topic belongs to.

    line 14

“Common” is measured: kept by at least 6 in 10 of the 7 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

Overview

Topic: $ARGUMENTS. This skill is the knowledge pipeline entry point: researcher, then fact-checker, then librarian, with on-disk hand-offs under research\.

When to use

You ask to research, learn or verify a topic, or an item from your research backlog is picked up. Not for a single quoted claim (use /verify "<claim>").

Steps

  1. Read CURRICULUM.md and the knowledge\ category indexes to pick the category the topic belongs to.
  2. Split the topic into 3 to 8 narrow questions, each answerable from one or two primary pages. Choose a <topic-slug>.
  3. Launch one researcher subagent per question in the same message so they run in parallel (/research-one does one). Each gets: the question, the topic slug, the question slug, and the source priority reminder. Expect back a path under research\inbox\<topic-slug>\.
  4. For each inbox file, run /verify research\inbox\<topic-slug>\<q>.md. Pass only the path, never the researcher's output.
  5. For each verified file, run /file research\verified\<topic-slug>\<q>.md.
  6. Reply: what was VERIFIED (with URLs), what was CONTRADICTED or UNVERIFIABLE, which notes changed (full paths), and the next topic. Mark questions NEEDS MAKER.

Common rationalizations

RationalizationReality
"I already know the answer, skip the researcher."Notes written from memory are where errors come from. Run the pipeline.
"Pass the researcher's reasoning to the fact-checker so it works faster."Independence is the point. Path only.
"The fact-checker said PARTIAL; file it and move on."PARTIAL is filed as the corrected value with the correction named.

Red flags

  • Fewer than 3 questions for a broad topic, or a question that needs opinion rather than a source.
  • Running researchers one at a time.
  • A note written by the lead instead of the librarian.

Verification

  • Every question has an inbox file, a verified file, and a librarian result.
  • tools\check_note.py PASS reported for every touched note.
  • Reply lists full paths and URLs.

Download now (ZIP)

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.