agentleFS
Sign inStart free
Boxline-devBoxline-dev/examples/skills/skill-tree/SKILL.md

skill-tree

Skill1 starsChanged 35 days agoMITView on GitHub

Turn a topic's official docs into a skill graph - 8 to 20 small linked concept notes plus an entry SKILL.md - that an agent can traverse instead of reading one long file. Searches and reads the docs with Boxline. Usage /skill-tree <topic-or-url>, add --global for ~/.claude/skills.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. jammyfu/open-game-skills★ 12same name
    2. maliklevraijeu/skill-tree★ 0same 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 repository4 other files it gives its agents.

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
  1. Distinctive:

    Every note links to at least one other, and is linked from at least one.

    line 41
  2. Distinctive:

    Only what the sources say.

    line 75

≈ “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

/skill-tree: a skill graph from docs

Build a graph of small, linked notes about one topic (a library, an API, a protocol) from its official docs.

Usage

/skill-tree <topic> or /skill-tree <docs-url>; --global saves to ~/.claude/skills/. Without input, show the usage (/skill-tree supabase-auth, /skill-tree https://docs.stripe.com/webhooks) and stop.

Boxline tools

The MCP server's web_search and web_fetch, or the CLI: npx @boxline/cli search "<query>" --limit 8 --json and npx @boxline/cli fetch <url>.

Steps

  1. Sources. One search: <topic> official documentation API reference. Take 3 or 4 official pages (docs site, official repository, official announcements); skip tutorials, aggregators and forums. With a URL as input, use it and add 1 or 2 official pages. With none credible, ask for a URL.
  2. Read. Fetch each as Markdown; skip one over 80,000 characters or one that fails. Keep the mechanisms, APIs, patterns, workflows, design decisions, security notes and pitfalls.
  3. Concepts. 8 to 20 atomic notes, each one of: a mechanism, a pattern, a constraint, a workflow, a design decision. No two notes on the same thing; split one that would pass 150 lines.

  4. Notes. One Markdown file per concept, 30 to 150 lines:

    ---
title: <Concept>
description: <one line an agent can read without opening the file>
links: [other-note, another-note]
---

<Prose with [[wikilinks]] inside sentences: "Policies read the user from [[jwt-claims]] …", not "Related: …">

    Every note links to at least one other, and is linked from at least one. Folders only for 3 or more notes that belong together.

  5. Entry point. SKILL.md, so the folder loads as a skill:

    ---
name: <topic>
description: <what the graph covers and when to use it>
---

# <Topic>: skill graph

<One paragraph on the domain.>

## Areas
- [[folder/]]: <what it covers>
  - [[note]]: <one line>
- [[note]]: <one line>

## Most connected
- [[note]] links <area> and <area>; most paths pass through it

## Gaps
- <what the sources did not cover>

## Sources
- <url> (read <YYYY-MM-DD>)
  6. Save to .claude/skills/<topic>/ (or ~/.claude/skills/<topic>/); ask before replacing an existing folder.
  7. Print the tree: <topic> (<n> notes, <m> links), then each folder and note with its links, and where it was saved.

Rules

  • Only what the sources say; no invented APIs.
  • At most 20 notes. Clear beats complete.
  • Public documentation only. No emojis.

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