A local code map that tracks functions, classes, imports, calls, and dependencies in a project. It can find meaningful code relationships instead of only matching text.

Use it to find callers and callees, search for symbols, understand a module, prepare context for editing, and assess the impact of a change. It requires an initial workspace index.

It reduces the need to search many files manually when investigating how code is connected or what a change might affect.

How codegraph-ai/codegraph-rules-for-agents configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to codegraph-ai/codegraph-rules-for-agents. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/codegraph-ai/codegraph-rules-for-agents/main/.claude/agents/codegraph.md

Clone the repo

git clone --depth 1 https://github.com/codegraph-ai/codegraph-rules-for-agents

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/agents/codegraph-ai/codegraph-rules-for-agents/codegraph/github.svg)](https://agentmods.dev/agents/codegraph-ai/codegraph-rules-for-agents/codegraph)

Your own site

<a href="https://agentmods.dev/agents/codegraph-ai/codegraph-rules-for-agents/codegraph"><img src="https://agentmods.dev/badge/agents/codegraph-ai/codegraph-rules-for-agents/codegraph/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/codegraph-ai/codegraph-rules-for-agents/codegraph"><img src="https://agentmods.dev/badge/agents/codegraph-ai/codegraph-rules-for-agents/codegraph.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 37 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,219 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00037	$0.01219
Opus 5.5	$0.00015	$0.00488
Sonnet 5.5	$0.00007	$0.00244
Haiku 4.5	$0.00004	$0.00122

Measured 2mo ago against content hash 2a02d296cbf4, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

codegraph scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/codegraph.md · 73 lines

How it starts

The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CodeGraph — local code intelligence

CodeGraph maintains a persistent semantic graph of your codebase (functions, classes, imports, call edges, dependency edges) in RocksDB. It resolves references — so "who calls X?" is a single tool call, not a multi-file grep.

First time in this project?

Run codegraph_reindex_workspace once. Takes 5-30 seconds. After that, the index persists across sessions automatically and updates incrementally.

Start here — the 5 tools you'll use 90% of the time

When you need	Call this	NOT this
Who calls function X?	`codegraph_get_callers(uri, line)`	grep for the function name
What breaks if I change X?	`codegraph_analyze_impact(uri, line)`	reading every importing file
Context for editing a file	`codegraph_get_edit_context(uri, line)`	reading 5+ files manually
Find a symbol by name	`codegraph_symbol_search(query)`	grep -r across the project
Module structure overview	`codegraph_get_module_summary(directory)`	ls + reading each file

These tools return resolved references (not string matches) — they know that auth::validate in file A calls jwt::verify in file B, even if the name doesn't appear as a literal string.

URI format: file:///absolute/path (not relative). Use the paths from symbol_search results directly.

Compact mode: most tools accept compact: true for shorter output. Use compact for scanning, full for deep investigation.

Common workflows (chain tools for best results)

PR review: pr_context — one call gives blast radius, test gaps, stale docs, reviewers
Refactoring: symbol_search → analyze_impact → get_edit_context
Bug triage: search_by_error → get_callers → get_ai_context(intent: "debug")
Onboarding: get_module_summary → find_entry_points → get_call_graph
Design check: index_markdown → verify_design → design_gaps

Decision rule

Before using Grep or reading multiple files for a code-structural question:

Read the full file on GitHub · 73 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 73 lines · 37 tokens per session scan A 2a02d296cbf4

Subscribe to this mod's changes

codegraph is an agent published in the GitHub repository codegraph-ai/codegraph-rules-for-agents (3 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 37 tokens to every session and 1,219 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Debugging

codegraph next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
codegraphcodegraph-ai/codegraph-rules-for-agents	Agent	3	5mo ago	Apache-2.0	A	37	Claude Code
review-obsidian-guidelinesuphy/obsidian-reminder	Agent	662	yesterday	MIT	A	63	Claude Code
debuggerTT-Wang/forge	Agent	35	3d ago	MIT	A	17	Claude Code
debuggerbostonaholic/rpikit	Agent	20	4d ago	MIT	A	28	Claude Code
loom-advisorcosmix/loom	Agent	65	6d ago	MIT	A	53	Claude Code
evolve-retrospectivemickeyyaya/evolve-loop	Agent	6	yesterday	Apache-2.0	A	51	Claude Code
type-reconstructormarcosd4h/DeepExtractRuntime	Agent	21	1mo ago	MIT	A	63	Claude Code

codegraph-rules-for-agents: Agent for Claude Code