Graphify knowledge graph
Project screenshotsAbout this Skill
Turn code and documents into a persistent knowledge graph for architecture questions and exploration. Graphify combines structural code analysis with semantic extraction to build a local graph, communities and reports. Its query, path and explain workflows help explore dependencies and project content; optional model-backed extraction has its own runtime and service requirements.
Why we picked it
Useful for repeated architecture questions and onboarding. A saved graph gives the agent a reusable map of relationships instead of starting each investigation with an unstructured repository tour.
What it helps with
- Turn code and documents into a persistent knowledge graph for architecture questions and exploration.
Setup and use
The first command installs the CLI and the second registers the assistant Skill. Use graphify install --platform codex for Codex, or --project for a project-scoped installation.
| Command | What it helps with |
|---|---|
uv tool install graphifyy && graphify install | Get the upstream package or workspace using the documented setup route. |
Upstream README and setupTask entrypoint and supporting files
Before you use it
- Python 3.10+ and a persistent uv or pipx installation. Install the assistant Skill after the CLI; keep its reference sidecar.
Supported agents
Sources & references
Original sources are linked below. Practical starting points are editorial suggestions from SkillHey.
