
Graft Code Graph Navigator
Lets an agent find, trace and map code in this repo via a precomputed graph instead of grepping files
- Repository
- trailhq/Trail-Code
- Stars
- 11,020
- Licence
- MIT
- Files
- 1 file
What it does
Once installed, this skill tells an AI coding agent to look up a prebuilt index of the repository (a set of markdown summaries plus a call/reference graph) before searching or reading source files. It offers six commands for answering questions like where code lives, what calls a function, and what a change might break, aiming to use fewer tokens and catch edges a plain text search would miss.
Great for
- Developers working in a large codebase who want an agent to answer 'where is X' or 'what calls Y' quickly
- Anyone doing a refactor or rename who needs to know every connected file before editing
- Teams wanting an architecture overview of an unfamiliar repo via graft map
Skip it if
- Repos that have not been indexed with graft: the commands depend on a graft/ directory and will not help without it
- Tasks needing up-to-date markdown docs under graft/ right after editing, since those cards can lag until the end of the turn; the skill itself warns against relying on them there
- Users expecting a no-setup experience: the skill assumes the graft CLI is installed and available to the agent
Try asking
- “How does authentication work in this repo? Use graft to find the relevant code.”
- “Before I rename the function parseConfig, show me everything that calls it.”
- “Give me an architecture overview of this codebase using graft map.”
- “Where is rate limiting handled, and what does that file expose?”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
graft.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip graft.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip graft.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Ask a specific conceptual or location question with --source for the most actionable results in one call.
- For refactors, run graft callers with --depth 2 or --depth all before editing to see the blast radius.
- If graft returns few or weak hits, switch to graft grep rather than rewording the same question.
What's inside
- SKILL.md
Redistributed unmodified from trailhq/Trail-Code under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.