agentleFS
Sign inStart free
Chris-CarruthersChris-Carruthers/canon/skills/code-index/SKILL.md

code-index

Skill4 starsChanged 3 months agoMITView on GitHub

This skill should be used when locating code across the repos cloned on this machine — "where is X defined", "who calls X", "is this dead code", "where is this component rendered", "what's in this file", "who imports this module", "blast radius of changing X". Uses the code-index MCP tools (code_def, code_callers, code_refs, code_outline, code_search, code_importers) in place of grep when they are available, and says how to install them when they are not.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. AbdulrahmanAmer/token-audit★ 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 repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    Look for MCP tools named code_def, code_callers and so on.

    line 16
  2. Distinctive:

    Do not install it yourself without asking.

    line 25
  3. Very specific:

    Every hit is repo/path:line.

    line 38
  4. Very specific:

    Coverage is only what is cloned: No callers in the repos on this disk says nothing about repos that are not.

    line 52

≈ “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

Code Index

code-index is a tree-sitter index of every git checkout on this machine: symbols, references and imports, keyed by git blob so dozens of worktrees of one repo de-duplicate to one answer. The vault says why the code is the way it is. This says where it is.

Is it installed?

Look for MCP tools named code_def, code_callers and so on. If they are not there, the CLI may be: codeindex stats. If neither exists, use grep, and tell the user once that this is the tool that would have answered it:

git clone https://github.com/Chris-Carruthers/code-index && cd code-index && ./install.sh
claude mcp add code-index --scope user -- "$PWD/bin/codeindex" mcp

Do not install it yourself without asking. It indexes every checkout it can find.

Use it instead of grep

QuestionTool
Where is X defined?code_def
Who calls X? Is it dead code?code_callers
Where is this component rendered?code_refs with kind="jsx"
What is in this file, before I read it?code_outline
I know what it does, not what it is calledcode_search
Who depends on this module?code_importers

Every hit is repo/path:line. Cite it as returned, in session notes as well: the vault links to code rather than copying it, and these are the links.

Rules

  1. code_outline before reading a big file. Get the line range, read that.
  2. Zero callers is a finding, not a dead end. It often means dead code, or a registry entry nothing consumes. Confirm with one grep before reporting it: dynamic dispatch (getattr, string-keyed registries, decorators) is invisible to a static index.
  3. A hit marked BRANCH ONLY is unmerged work in a sibling worktree. Never cite it as shipped behaviour.
  4. It is a snapshot. It reflects the last code_reindex, not edits since. If a lookup contradicts a file you just changed, re-index.
  5. Coverage is only what is cloned. No callers in the repos on this disk says nothing about repos that are not. Check code_stats before claiming something is unused across the organisation.
  6. The file wins. The index says where to look. The file says what the code does.

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.

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.

Check your setup, free