⚡ TimmyTest

Zero-Token Test Runner • AST Test Gap Analyzer • AI Agent Prompt Generator

Run tests locally. Find likely missing tests. Hand your coding agent a concise report.

TimmyTest terminal demo: project overview, test results, failure diagnosis, test gaps and AI agent handoff prompt

Try TimmyTest View source on GitHub Quick Start AI agent testing guide

Why TimmyTest?

TimmyTest detects the test setup, runs your existing suite and turns failures and likely test gaps into a short report for your AI coding agent. Its local commands make no AI API calls; sending the resulting prompt to an agent still uses that agent's context and tokens.

⚡ 0 AI tokensTest execution, gap analysis and diagnostics run 100% locally.

🔍 AST gap analysisEvery source module mapped to its tests; missing files flagged by priority.

🩺 Root-cause fixesRule-based suggestions from real tracebacks — expected vs actual.

🌍 35+ ecosystemsPython, JS/TS, Rust, Go, Java, .NET, Dart, Solidity and 27 more.

🔌 MCP serverNative tools for Claude Desktop/Code, Cursor, Zed, Windsurf.

🛡️ CI exit codesAssertion failures AND unloadable suites break the pipeline.

Quick start

From a terminal in a project that already has tests:

python -m pip install "git+https://github.com/tugrakaymakcioglu/TimmyTest.git@v2.0.1"
timmytest check .

The package is currently installed from its public GitHub source. PyPI publication is pending. See the demo and example output before trying it on your own project.

If installation fails or the report is unclear, tell us what happened with your OS, command and a redacted output sample.

Testing during vibe coding

When an AI coding agent changes a repository, TimmyTest can run the existing tests and give the agent a smaller failure report before it starts debugging. Its agent testing guide shows when to choose the CLI, when to use the MCP tools, and when to fall back to the project's own runner.