Mutation testing for Python, written in Rust. Injects faults into Python source code and runs the test suite to measure how well tests catch real bugs. Built for speed — parses and mutates Python ASTs natively in Rust, then shells out to the Python test runner.

Static call graph generation for Python, in Rust. Produces whole-program call graphs without executing code. Useful for understanding large codebases, dead code detection, and as an input to other static analysis tools.

Control flow graph generation for Python, in Rust. Builds per-function CFGs from Python source, handling branches, loops, exception handlers, and context managers. Outputs DOT format for visualization or further analysis.

Extracts DeepWiki documentation into a single LLM-ready markdown file. Crawls a DeepWiki-generated site and consolidates all pages into one document, preserving structure and code blocks for use as context in language model prompts.

Multi-agent coding orchestrator. Coordinates multiple AI coding agents working on the same codebase, handling task decomposition, context sharing, and conflict resolution across parallel workstreams.

Minimal coding agent harness. A lightweight framework for building AI coding agents with tool use, context management, and conversation loops. Designed to be small enough to understand in one sitting.

Multiplayer AI crisis simulation game. Players take on roles in a near-future scenario where rapid AI capability gains force real-time strategic decisions. Built as a web app with real-time multiplayer via WebSockets.

Detects AI writing patterns in prose. Scans text for common LLM stylistic tells — hedging phrases, filler transitions, over-qualification, and other patterns that mark machine-generated text. Written in Rust for use as a CLI tool or library.