AGENT SKILL

graphify

Use when navigating an unfamiliar codebase, tracing how code connects (callers, dependencies, the blast radius of a change), or whenever the repo ships a graphify-out/graph.json knowledge graph. Query the graph before grepping/reading raw source for orientation.

License 许可
MIT
最近更新
2026-11-09
Stars
★ 6
Forks
1

01 前言 Front Matter

name
graphify
license
MIT
metadata
{ "wraps": "Graphify — Apache-2.0, (c) 2026 Safi Shamsi and the Graphify contributors", "author": "ai-playbook", "notice": "../../NOTICE", "version": "1.0" }
description
Use when navigating an unfamiliar codebase, tracing how code connects (callers, dependencies, the blast radius of a change), or whenever the repo ships a graphify-out/graph.json knowledge graph. Query the graph before grepping/reading raw source for orientation.

03 SKILL.md 正文

graphify — knowledge-graph code navigation

graphify (safishamsi/graphify, PyPI graphifyy) parses a repo into a structured knowledge graph — nodes are files/functions/classes, edges are imports/calls/inheritance — plus community clustering and a semantic label layer. Querying the graph returns a small, structurally-precise subgraph instead of a pile of raw file reads, so orientation costs far fewer tokens and round-trips.

When to use

Act only when graphify-out/graph.json exists in the repo. Reach for graphify FIRST (before grep/read) when the task is orientation or structure:

  • "How does <subsystem> work / where is <concern> handled?"
  • "Who calls / what depends on <symbol>?" — blast radius before a change.
  • "What is the path between <A> and <B>?"
  • Onboarding to an unfamiliar area of the repo.

If graphify-out/graph.json is absent, graphify is not set up here — fall back to normal search.

Commands

  • graphify query "<question>" — scoped subgraph for a natural-language question. The default entry point.
  • graphify explain "<concept>" — focused view of one concept/community.
  • graphify path "<A>" "<B>" — the relationship chain between two symbols.
  • graphify update . — re-parse changed files into the graph (AST-only, no API/token cost). Run after editing code so the graph stays fresh.
  • graphify-out/wiki/index.md, if present, is good for broad navigation.
  • graphify-out/GRAPH_REPORT.md — read only for a broad architecture pass when query/path/explain do not surface enough.

Discipline

  • Query-first. On a code-exploration task, run graphify query before raw grep/read. Grep AFTER the graph has oriented you, or to modify/debug specific lines.
  • Pass it down. Include the query-first instruction in any subagent prompt that explores code.
  • Update-after-edit. After changing code, run graphify update . so the next session's map is correct. A stale graph misleads.

When NOT to use

  • Editing a file you already know — go straight to it; graphify adds overhead.
  • Prose/document Q&A with no code structure — graphify is structural, not a text-similarity RAG.
  • The graph is stale or absent — regenerate (graphify update .) or fall back to grep; never trust a stale map.

Boundaries

  • graphify is a navigation aid, not a source of truth — verify a claim against the actual code before asserting it (the graph can lag the working tree).
  • It complements, not replaces, grep/read for line-level work.

Setup

Once per clone: uv tool install "graphifyy>=0.8.31" (or pipx) + graphify hook install. See docs/runbooks/graphify-setup.md. The graphify-adoption rule enforces the gitignore + merge-driver hygiene that keeps the committed graph portable across machines.

See also

04 同仓库更多 Skills

Wizarck
Wizarck/ai-playbook

sweep

Use when the user wants to find code that has quietly stopped being used — types /sweep, asks for a "dead code sweep", "repo entropy", "what can we delete", "orphan files", "unused code", or asks to review a sweep ledger. Adjudicates candidates a deterministic scanner already produced; it never deletes anything and never re-derives findings from raw source. Pairs with ponytail, which gates what gets written rather than what rots.

★ 6
Wizarck
Wizarck/ai-playbook

ponytail

Use when the user wants the laziest solution that actually works — types /ponytail, says "be lazy", "lazy mode", "simplest solution", "minimal solution", "yagni", "do less", or "shortest path", or complains about over-engineering, bloat, boilerplate, or unnecessary dependencies. Supports intensity levels lite, full, ultra (default). Pairs with caveman, which compresses prose, not code.

★ 6
Wizarck
Wizarck/ai-playbook

ponytail-review

Use when the user wants a code review focused exclusively on over-engineering — says "review for over-engineering", "what can we delete", "is this over-engineered", "simplify review", or invokes /ponytail-review. Complements correctness/security review; this one only hunts complexity in the current diff. Gated by the ponytail review_ponytail component.

代码审查
★ 6
Wizarck
Wizarck/ai-playbook

ponytail-debt

Use when the user wants to harvest every `ponytail:` comment in the codebase into a tracked debt ledger — says "ponytail debt", "/ponytail-debt", "what did ponytail defer", "list the shortcuts", "ponytail ledger", or "what did we mark to do later". One-shot report, changes nothing. Gated by the ponytail debt_ponytail component.

★ 6
Wizarck
Wizarck/ai-playbook

ponytail-audit

Use when the user wants a whole-repo audit for over-engineering — says "audit this codebase", "audit for over-engineering", "what can I delete from this repo", "find bloat", or invokes /ponytail-audit. Like ponytail-review but scans the entire tree, not a diff. One-shot report, applies no fixes. Gated by the ponytail audit_ponytail component.

★ 6
Wizarck
Wizarck/ai-playbook

openspec-propose

Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.

★ 6