graphify skill
The /graphify skill builds a knowledge graph of a project's code, docs and PDFs so the agent queries the graph instead of grepping
- 138,535 ★ on the repository
- Last commit
- Apache-2.0
Extended access
Why this level
- Reads all project files and writes results to graphify-out
- Docs and media are sent to a model for analysis
- Installs git hooks on a separate command
Install
- Works in
Manual install
uv tool install graphifyy
graphify installAlternatives to uv are pipx install graphifyy or pip install graphifyy. Add --project to install the skill into the current repository.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Reads all project files and writes results to graphify-out
- Docs and media are sent to a model for analysis
- Installs git hooks on a separate command
This is third-party code. Review the repository files before installing.
What it does
The /graphify command walks a folder or repository and builds a graph of links between files, functions, schemas and documents. Code is parsed locally with tree-sitter without a model, while docs, PDFs, images and video get a semantic pass from the agent's model. Every edge is tagged as extracted from source or inferred, so you can tell what to trust. Outputs are interactive HTML, GraphRAG-ready JSON and a GRAPH_REPORT.md, and the agent gets query, path and explain commands plus an optional MCP server over the graph.
Who it is for. For developers and teams using agents in large or unfamiliar codebases.
Good fit when
- You need to understand the architecture of an unfamiliar repository quickly
- The agent burns many tokens searching through files
- You want one map linking code with docs, SQL schemas and PDFs
Not a fit when
- A small project the agent can read in full anyway
- Documents cannot be sent to a model but document analysis is what you need
Example request
/graphify . then show how the auth module connects to the databaseLimitations
Requires Python and the graphifyy package from PyPI (double y; other graphify packages are unrelated). Code parsing is local, while docs, images and video go through the agent's model or a separate API key. The graph must be refreshed after pulls; git hooks help with that.
How to disable. Remove the skill from the agent's skills directory and the package with uv tool uninstall graphifyy or pip uninstall graphifyy. Delete the graphify-out folder in the project manually if needed.
FAQ
Does code leave the machine?
No, code is parsed locally via AST. Only docs, PDFs and media you include go to a model.
Can a team share one graph?
Yes, graphify-out is meant to be committed, and the built-in MCP server can serve the graph over HTTP.
How is graphify different from grepping code?
The /graphify command builds a graph of links between files, functions, schemas and documents. The agent answers with graph queries instead of scanning files. Code is parsed locally with tree-sitter, while documents and PDFs get semantic analysis by the agent model.
How do I install the graphify skill for Claude Code and Cursor?
Install with uv: uv tool install graphifyy, then graphify install. The /graphify command works in Claude Code, Cursor, Codex and Gemini CLI and builds the graph for a folder or repository.
Discussion
No questions or reviews yet.
Related
SkillsCode graph and project memory in one Rust program: the agent gets excerpts with file and line citations
An MCP server and CLI that build a knowledge graph of a monorepo and let an agent query, understand and edit multi-language code
CodeGraph (Rust)
codegraph-rust
A Rust code graph for agents: four MCP tools show what an edit will touch, how the architecture fits together and where the fragile spots are
Local code graph for agents: an MCP server returns entry points, symbol links and change impact in one call
Indexes a codebase into a knowledge graph and gives the agent call chains, change blast radius and code search over MCP
Local code structure graph: builds a repository map with tree-sitter and serves the assistant only the context a change actually touches over MCP
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| graphify (this entry) | no | 138,535 | ||
| Ix | no | 1,137 | ix turns a repository into a persistent graph of symbols | |
| SwarmVault | yes | 731 | a CLI and MCP server that builds a local wiki with a knowledge graph | |
| Graphicle: code graph and project memory for agents | no | not required | 432 | code graph and project memory in one Rust program |
README in short
The README covers a two-step install: the graphifyy package via uv, pipx or pip and registering the skill with graphify install, optionally per project. It describes team mode with git hooks that rebuild the graph on commit and branch switch, and an MCP server with query_graph, get_neighbors and shortest_path tools. Supports Claude Code, Cursor, Codex, Gemini CLI and more. Apache-2.0 licensed.
README badge
Original SKILL.md
--- name: graphify description: "Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community detection, and query/path/explain tools." --- # /graphify Turn any folder of files into a navigable knowledge graph with community detection, an honest audit trail, and three outputs: interactive HTML, GraphRAG-ready JSON, and a plain-language GRAPH_REPORT.md. ## Usage ``` /graphify # full pipeline on current directory (HTML viz; add --obsidian for a vault) /graphify <path> # full pipeline on specific path /graphify https://github.com/<owner>/<repo> # clone repo then run full pipeline on it /graphify https://github.com/<owner>/<repo> --branch <branch> # clone a specific branch /graphify <url1> <url2> ... # clone multiple repos, build each, merge into one cross-repo graph /graphify <path> --mode deep # thorough extraction, richer INFERRED edges /graphify <path> --update # incremental - re-extract only new/changed files /graphify <path> --directed # build directed graph (preserves edge direction: source→target) /graphify <path> --whisper-model medium # use a larger Whisper model for better transcription accuracy /graphify <path> --cluster-only # rerun clustering on existing graph /graphify <path> --no-viz # skip visualization, just report + JSON /graphify <path> --html # (HTML is generated by default - this flag is a no-op) /graphify <path> --svg # also export graph.svg (embeds in Notion, GitHub) /graphify <path> --graphml # export graph.graphml (Gephi, yEd) /graphify <path> --neo4j # generate graphify-out/cypher.txt for Neo4j /graphify <path> --neo4j-push bolt://localhost:7687 # push directly to Neo4j /graphify <path> --falkordb # generate graphify-out/cypher.txt for FalkorDB /graphify <path> --falkordb-push falkordb://localhost:6379 # push directly to FalkorDB