agentleFS
Sign inStart free
RandyRawgustRandyRawgust/ai-knowledge-base-template/.claude/skills/compile-topic/SKILL.md

compile-topic

Skill1 starsChanged 6 months agoMITView on GitHub

Invoke topic-compiler specialist to synthesize a Karpathy-style topic note from raw sources in 11 - Sources/. Use when Andy says compile a topic, build a topic note, summarize sources on X, make a primer on Y, what do I know about Z.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    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
  1. Distinctive:

    Read every source. Extract key claims, definitions, examples, contradictions.

    line 27
  2. Distinctive:

    Don't write from training knowledge alone - this is source-grounded synthesis

    line 55
  3. Distinctive:

    Don't skip contradictions between sources - surface them, don't paper over

    line 56
  4. Distinctive:

    Don't make it longer than the sources warrant - density beats length

    line 57
  5. Distinctive:

    Don't auto-tag - tags are Andy's call

    line 58

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

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

Compile Topic

Wrapper for [[03 - Skills & Rules/Agents/Specialists/topic-compiler|topic-compiler specialist]]. Builds a coherent topic note in 10 - Topics/ from raw clipped sources.

When to use

Triggers: compile a topic, build a topic note, summarize sources on X, make a primer on Y, what do I know about Z, synthesize these sources.

Load

  • 03 - Skills & Rules/Agents/Specialists/topic-compiler.md - canonical spec
  • 11 - Sources/ - raw inputs (Web Clipper destination)
  • An existing topic note as style reference: pick any from 10 - Topics/AI/ or related domain
  • 03 - Skills & Rules/Agents/memory/Archivist/universal.md - Karpathy-wiki tone

Confirm inputs

  1. Topic name (becomes file slug)
  2. Source list - paths, a glob like 11 - Sources/2026-05-*, or "everything tagged X"
  3. Audience - just-me notes vs. shareable explainer
  4. Domain folder under 10 - Topics/ (AI, Game Dev, etc.) - create if doesn't exist

Run

  1. Flip topic-compiler to status: active
  2. Read every source. Extract key claims, definitions, examples, contradictions.
  3. Group by sub-theme. Identify the spine of the topic (3-5 main sections).
  4. Draft the note in Karpathy style: dense, linked, definitions inline, examples concrete, no fluff.
  5. Cite each source via [[11 - Sources/<file>]] wikilinks. Every non-obvious claim cites.

Output

10 - Topics/<Domain>/<topic-slug>.md with frontmatter:

---
type: topic
created: <MM-DD-YY>
domain: <Domain>
sources: [<list of source wikilinks>]
status: draft | reviewed
tags: [topic, <domain-tag>]
---

Body sections: intro, main concepts (each subsectioned), open questions, related topics, sources.

After

  1. Cross-link from related existing topic notes
  2. Append Activity Log row
  3. Flip topic-compiler to status: idle
  4. Provide [Open topic note](computer://...) link

Don't

  • Don't write from training knowledge alone - this is source-grounded synthesis
  • Don't skip contradictions between sources - surface them, don't paper over
  • Don't make it longer than the sources warrant - density beats length
  • Don't auto-tag - tags are Andy's call

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.

Check your setup, free