graphify

A command and code knowledge graph for exploring how a project is structured. It reads Python syntax to connect files, functions, modules, calls, and dependencies.

In plain words
What is it for?
Use it to find what calls a function, trace data flow, inspect dependencies, understand architecture, or locate important code nodes.
Why use it?
It gives a direct way to answer structural code questions without manually searching many files. The graph can also be rebuilt when the code changes.

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/fwornle/coding/graphify
Clone the repo
git clone --depth 1 https://github.com/fwornle/coding

Made for: Claude Code.

Per session 80 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 682 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 $0.00080 $0.00682
Opus 5 $0.00040 $0.00341
Sonnet 5 $0.00016 $0.00136
Haiku 4.5 $0.00008 $0.00068

Measured yesterday against content hash 0d0879ad4e37, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

graphify 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 yesterday.

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/graphify.md · 55 lines

How it starts

The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/graphify

Graphify turns this repo into a navigable code knowledge graph (tree-sitter AST → static graph.json), served over an HTTP MCP endpoint. All Python runs inside the coding-services container — the host graphify command (bin/graphify) forwards to it via docker exec.

  • Graph output: .data/graphify/graphify-out/graph.json (bind-mounted; built_at_commit stamps the indexed commit)
  • MCP endpoint: http://localhost:3851/mcp (tools: query_graph, get_node, get_neighbors, shortest_path, graph_stats, god_nodes, …)
  • Host CLI: graphify … (shim → container)

When to use

Prefer this over grepping the codebase for structural questions ("how does X work?", "what calls Y?", "trace the flow through Z", "what depends on this module?"). Query the graph first; fall back to grep only if the graph lacks the answer.

Querying (fast path — graph already built)

Use the MCP tools (mcp__graphify__query_graph, get_node, get_neighbors, shortest_path, god_nodes) when available. Equivalent CLI via the shim:

graphify query "How does the ETM watchdog reclaim a stalled session?"   # BFS, broad context
graphify query "what calls captureForegroundTokens" --dfs               # DFS, trace a path
graphify path "ObservationWriter" "obs-api"                             # shortest path between two concepts
graphify explain "CodeGraphAgent"                                       # plain-language node explanation
graphify god-nodes --top 20                                            # most-connected hubs

Rebuilding the graph

The dashboard shows how many commits behind the graph is and has a Re-index button. To rebuild from the CLI:

graphify update /workspace/coding        # incremental (AST only, no LLM) — fast, use this most of the time
graphify extract /workspace/coding       # full re-extract incl. docs/PDF semantic pass (routes docs LLM via the proxy)
graphify extract /workspace/coding --code-only   # full code re-extract, no LLM/network

Read the full file on GitHub · 55 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. yesterday First seen · 55 lines · 80 tokens per session scan A 0d0879ad4e37

Subscribe to this mod's changes

graphify is a command published in the GitHub repository fwornle/coding (2 stars, last pushed yesterday), licensed MIT. It adds 80 tokens to every session and 682 once invoked, about $0.0004 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 2026-08-31.