graphifyy

What it is and what it does

Graphifyy is a CLI tool that transforms a folder of source code, documentation, images, and video into an interactive knowledge graph you can query instead of searching files. It uses tree-sitter to parse code deterministically into an AST—no LLM involved, nothing leaves your machine—and optionally adds semantic analysis of non-code media via your AI assistant's model or a configured backend. The result is three files: an interactive HTML graph, a markdown report of key concepts and connections, and a JSON graph you can query later.

The package integrates as a skill into popular AI coding assistants (Claude Code, Cursor, Codex, Gemini CLI, and others) so you can invoke it with a slash command. Every edge in the graph is tagged as either EXTRACTED (explicit in source code) or INFERRED (resolved by graphify), so you know what was read directly versus inferred. It detects communities within the graph, identifies highly-connected nodes, and resolves cross-file relationships like calls, imports, inheritance, and mixins across multiple languages.

Use it for:

• Map a large unfamiliar codebase to understand its architecture, key concepts, and how components connect without manually reading files.
• Query a knowledge graph to find how two unrelated functions or classes are connected, or trace the path between them.
• Generate a markdown report of surprising connections and suggested questions about a project's structure for onboarding or documentation.
• Index code, docs, PDFs, and images together so you can ask questions that span multiple media types within a single graph.
• Commit a project-scoped graph to version control so teammates can query the same knowledge graph without re-parsing the codebase.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Parses code, docs, PDFs, images, and video into a queryable knowledge graph for AI coding assistants, using tree-sitter AST for deterministic code analysis and optional semantic passes for other media.

Yes, if you use an AI coding assistant and want to explore or understand a codebase faster than grepping or reading files. The local-first code parsing, permissive license, and active maintenance make it low-risk. Install friction is minimal (pure Python wheel, standard dependencies). No security vulnerabilities reported. Start with a small project to verify the graph quality matches your needs.

Install

graphifyy on PyPI

pip install graphifyy

uv add graphifyy

poetry add graphifyy

Installing graphifyy

Before you install

Low friction install as a pure Python wheel. Active maintenance with recent release. Substantial runtime dependency footprint (29 packages including tree-sitter parsers for multiple languages), but all are standard PyPI packages with no compiled system dependencies.

License in practice

Apache-2.0 permissive license allows commercial and private use without restriction, modification, or redistribution obligations beyond license inclusion.

Quickstart

# Install
uv tool install graphifyy

# Register with AI assistant
graphify install

# In your AI assistant, run:
# /graphify .

# Outputs appear in graphify-out/:
# - graph.html (interactive browser view)
# - GRAPH_REPORT.md (highlights and suggested queries)
# - graph.json (full queryable graph)

Requires Python 3.10 or later. Code parsing is local and LLM-free, but semantic passes over docs/media require either your AI assistant's model or a configured API key.

Verify before relying

• Whether the 29 tree-sitter language parsers are all installed by default or selectively based on detected project languages.
• Performance characteristics and graph size limits for very large codebases.
• Whether graph.json can be queried programmatically via Python API or only through CLI commands (explain, path, query).

Package facts

Evidence: graphifyy-0.9.43-py3-none-any.whl

Keywords: claude, claude-code, codex, opencode, kilo, cursor, gemini, aider, kiro, pi, devin, knowledge-graph, rag, graphrag, obsidian, community-detection, tree-sitter, leiden, llm

Tags

More Software Development packages

Further reading