graphify
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini
What is graphify?
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.
Overview of graphify
What it is
graphify is a tool that converts a codebase—including its documentation, SQL schemas, configuration files, and PDFs—into a queryable knowledge graph. It operates as a /graphify skill for AI coding assistants such as Claude Code, Cursor, Codex, and Gemini CLI. Code is parsed locally using tree-sitter AST without any LLM involvement or data leaving the machine; documentation and media use a configured API key for a semantic pass. The graph is not a vector index but a real graph that users traverse by asking questions, tracing paths between concepts, or explaining a single concept. Each edge carries a tag—EXTRACTED (explicit in the source) or INFERRED (resolved by graphify)—to indicate how the connection was derived. The tool produces three files: graph.html (browser-interactive graph), GRAPH_REPORT.md (key concepts and suggested questions), and graph.json (the full graph for later queries). The persisted graph includes graph.schema_version and graph.graphify_version for compatibility detection. Graphify Cloud, a separate always-on service, offers always-on memory across the SDLC with formal verification, cross-repo support, and connectors for Sentry and Jira.
Key features
- Parses code locally with tree-sitter AST: deterministic, no LLM, nothing leaves the machine.
- Every edge is tagged
EXTRACTED(explicit in source) orINFERRED(resolved by graphify). - No embeddings or vector store; a real graph that users traverse.
- Supports queries, path tracing between two things, and explaining a single concept.
- Works with 20+ AI coding assistants including Claude Code, Cursor, Codex, Gemini CLI, and GitHub Copilot.
- Handles code, docs, PDFs, images, and video/audio into the same graph.
Use cases
- Querying a codebase instead of reading files: ask a plain-language question and get a scoped subgraph.
- Tracing how any two components are connected via
graphify path. - Explaining a single concept (e.g.,
APIRouter) with all its connections, including confidence tags. - Understanding project structure across multiple repos or monorepos—Graphify Cloud extends this with cross-repo links and SDLC-wide code review.
Getting started
bash uv tool install graphifyy # install the CLI (or: pipx install graphifyy) graphify install # register the skill with your AI assistant Then, in your AI assistant: /graphify .
License and pricing
The license is Apache-2.0. The project is open source (pricing_type: open_source). Graphify Cloud, a separate service, offers a free 14-day trial—details for that service are available at app.graphify.com.
Based on the project's public README and repository metadata.