graph

A read-only command for asking Cypher questions of a local Neo4j code graph. A code graph is a searchable map of files, classes, functions, methods, imports, and other relationships in a codebase.

In plain words
What is it for?
Use it to find who imports a file or symbol, locate classes, inspect methods, and measure dependencies or other relationships in the project.
Why use it?
It lets you inspect how code is connected without manually searching through every file. The command assumes the local graph database is running and has been indexed.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/cognitx-leyton/codegraph/graph
Clone the repo
git clone --depth 1 https://github.com/cognitx-leyton/codegraph

Made for: Claude Code.

Per session 16 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 703 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00016 $0.00703
Opus 5 $0.00008 $0.00351
Sonnet 5 $0.00003 $0.00141
Haiku 4.5 $0.00002 $0.00070

Measured 2d ago against content hash c5336995e375, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

graph 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 2d 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/commands/graph.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Usage

/graph <cypher>

Runs the Cypher against the local Neo4j instance via codegraph query --json. Read-only — writes (CREATE, MERGE, DELETE, SET) are not rejected at the CLI level (unlike the MCP server), so be careful what you execute. If you're unsure, stick to MATCH queries.

Assumes:

  • Neo4j is up at bolt://localhost:7688 (docker compose up -d from codegraph/ if not).
  • The graph has been indexed at least once (run /graph-refresh if you get zero results on queries you expect to match).

Canonical query patterns for this repo

~20 files, 56 classes, 134 module functions, ~180 methods

Blast radius — who depends on a file?

MATCH (f:File)-[:IMPORTS]->(target:File {path: 'codegraph/codegraph/schema.py'})
RETURN f.path AS caller
ORDER BY f.path

Who imports a specific symbol?

MATCH (f:File)-[r:IMPORTS_SYMBOL]->(g:File)
WHERE r.symbol = 'IgnoreFilter'
RETURN f.path AS caller, g.path AS source

Find a class by substring

MATCH (c:Class)
WHERE c.name CONTAINS 'Node' AND c.file STARTS WITH 'codegraph/codegraph/'
RETURN c.name, c.file
ORDER BY c.name

Methods on a class

MATCH (c:Class {name: 'IgnoreFilter'})-[:HAS_METHOD]->(m:Method)
RETURN m.name, m.visibility, m.is_constructor
ORDER BY m.name

Census of dataclasses

MATCH (f:File)-[:DEFINES_CLASS]->(c:Class)-[:DECORATED_BY]->(d:Decorator {name: 'dataclass'})
WHERE f.path STARTS WITH 'codegraph/codegraph/'
RETURN f.path, count(c) AS dataclass_count
ORDER BY dataclass_count DESC

What decorators does a file use?

MATCH (:File {path: 'codegraph/codegraph/mcp.py'})-[:DEFINES_FUNC]->(fn:Function)-[:DECORATED_BY]->(d:Decorator)
RETURN fn.name, d.name
ORDER BY fn.name

Unresolved (external) imports from a file

MATCH (:File {path: 'codegraph/codegraph/cli.py'})-[r:IMPORTS {external: true}]->(e:External)
RETURN e.specifier
ORDER BY e.specifier

Read the full file on GitHub · 91 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.

  1. 2d ago First seen · 91 lines · 16 tokens per session scan A c5336995e375

Subscribe to this mod's changes

graph is a command published in the GitHub repository cognitx-leyton/codegraph (11 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 16 tokens to every session and 703 once invoked, about $0.0001 per session on Opus 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 2026-08-30.