graphmind CLAUDE.md

Project instructions for Graphmind, a local code-intelligence command-line tool that parses code, stores relationships in SQLite, and connects to coding agents through MCP.

In plain words
What is it for?
Building, checking, linting, testing, setting up, and initializing the Graphmind Rust workspace.
Why use it?
They give an agent the project’s technology choices, commands, setup steps, and folder layout so it can work consistently in the repository.

Instructions file

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 instructions/aouicher/graphmind/claude-md
Clone the repo
git clone --depth 1 https://github.com/aouicher/graphmind
Per session 811 This file is loaded in full into every session.
When invoked 811 The same file — it is already loaded in full.
Security scan C 2 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 $0.00811 $0.00811
Opus 5 $0.00405 $0.00405
Sonnet 5 $0.00162 $0.00162
Haiku 4.5 $0.00081 $0.00081

Measured 3d ago against content hash bc44341cbd74, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

graphmind CLAUDE.md scanned grade C with 2 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 3d 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

curl -fsSL https://raw.githubusercontent.com/aouicher/graphmind/main/scripts/install.sh | bash

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -fsSL https://raw.githubusercontent.com/aouicher/graphmind/main/scripts/install.sh | bash
CLAUDE.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

graphmind

Local-first code intelligence CLI with MCP server.

Tech Stack

  • Rust — full Cargo workspace
  • Parsing: tree-sitter via napi-rs (crates/graphmind-core/)
  • Graph: SQLite + FTS5 (crates/graphmind-db/)
  • Memory: JSONL store + cross-project inference (crates/graphmind-memory/)
  • Embeddings: cosine search + RRF (crates/graphmind-embeddings/)
  • MCP: rmcp SDK v1.5, stdio transport (crates/graphmind-mcp/)
  • CLI: clap (crates/graphmind-cli/)

Commands

cargo check --workspace       # check all crates
cargo build --release -p graphmind-cli  # build release binary
cargo clippy --workspace -- -D warnings # lint
cargo test --workspace        # run tests

User-facing setup

graphmind setup               # global one-time: hooks + MCP + skill
graphmind init                # per-project: register + git hooks + build
graphmind init --skip-build   # per-project without building

Project Structure

  • crates/graphmind-core/ — Rust parsing engine (tree-sitter, napi-rs)
  • crates/graphmind-db/ — SQLite graph (schema, builder, queries, FTS5, cache)
  • crates/graphmind-memory/ — JSONL memory store + cross-project links
  • crates/graphmind-embeddings/ — embedding store, cosine search, RRF fusion
  • crates/graphmind-mcp/ — MCP server (rmcp SDK, 24 tools)
  • crates/graphmind-cli/ — CLI (clap, all commands)

Installation

curl -fsSL https://raw.githubusercontent.com/aouicher/graphmind/main/scripts/install.sh | bash

graphmind

Last build: 2026-05-01 | 934 symbols | 2246 edges | 131 files Languages: rust (92), typescript (20), toml (9), markdown (3), bash (2), ruby (1), python (1), html (1), go (1), css (1) MCP: graphmind mcp (stdio)

MANDATORY: use graphmind for ALL code exploration

NEVER use grep/find/ls/Glob to search code. Use graphmind MCP tools:

Need MCP tool CLI equivalent
Find symbol gm_fn graphmind fn <name>
Search by intent gm_search graphmind search "<query>"
File dependencies gm_deps graphmind deps <file>
Symbol resolution gm_query graphmind query <name>
Blast radius gm_fn_impact graphmind fn-impact <name>
Git diff impact gm_diff_impact graphmind diff-impact
Project overview gm_map graphmind map
File outline gm_outline graphmind outline <file>
Who calls chain gm_who_calls_chain
Dead code gm_dead_code
Cross-project gm_cross_query graphmind cross query
Memory search gm_memory_search graphmind memory search

Read the full file on GitHub · 74 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. 3d ago First seen · 74 lines · 811 tokens per session scan C bc44341cbd74

Subscribe to this mod's changes

graphmind CLAUDE.md is an instructions file published in the GitHub repository aouicher/graphmind (207 stars, last pushed 1mo ago), licensed MIT. It adds 811 tokens to every session, about $0.0041 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.