القائمة
Local symbol-and-relationship map of the repo. Use to see who calls what (blast radius) before editing a function, class, or method. The binary is at /arm/bin/cg.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | codegraph |
| description | Local symbol-and-relationship map of the repo. Use to see who calls what (blast radius) before editing a function, class, or method. The binary is at /arm/bin/cg. |
codegraph builds a function-level dependency graph of the whole repo
(tree-sitter) and lets you query it. In this environment the binary is at
/arm/bin/cg — call it by full path.
/arm/bin/cg build .
Creates .codegraph/graph.db. Takes ~1-2s for most repos. Idempotent.
Before changing a symbol, check its blast radius.
| Question | Command |
|---|---|
| What symbols are in this file, and how many callers does each have? | /arm/bin/cg brief <file> |
| Who calls this symbol (callers / fan-in, tests excluded)? | /arm/bin/cg where <SymbolName> -T |
| What breaks transitively if this symbol changes? | /arm/bin/cg fn-impact <SymbolName> |
| Full context: source + deps + callers + tests? | /arm/bin/cg context <SymbolName> |
| What does this file import / what imports it? | /arm/bin/cg deps <file> |
brief is the cheap first look (one line per symbol with a caller count and a
risk tier like [utility, 2 callers] or [core, 7 callers]). where/fn-impact
give you the actual caller names when a count looks dangerous.
brief tags each symbol: core (heavily depended on), utility, leaf
(no callers), dead-ffi/dead. A core symbol with many callers is high blast
radius — changes ripple widely.
where to see
callers, or brief on the file for a quick scan.fn-impact first.CodeGraph scout before broad grep/read. Use for repo explanation, navigation, diagnosis, runtime/reconnect flow, contract/RPC/schema tracing, refactor/cycle seams, dead-code cleanup, test targeting, and code review.
CodeGraph scout before broad grep/read. Use for repo explanation, navigation, diagnosis, runtime/reconnect flow, contract/RPC/schema tracing, refactor/cycle seams, dead-code cleanup, test targeting, and code review.
Paired trajectory analysis for benchmark churn. Use when comparing two configs on matched task/rep cells, explaining solve flips, diagnosing why one config solved and another failed, separating net score from churn, or preparing evidence to improve a skill/prompt/tool from trajectory differences.
Prompt embedding analysis. Use when clustering benchmark prompt/config text, comparing semantic neighbors, or separating prompt-shaped effects from behavioral wrappers in deep-swe-bench results.
Use before launching harness/run_batch.py for a benchmark, especially when configs use advisor, observational-memory workers, subagents, local-vllm shims, or any model beyond the main executor; use before claiming a benchmark launch is working.
Use before adding or changing a deep-swe-bench config, model leaf, provider/model API path, usage parser, smoke contract, or extension/subagent worker usage accounting.