Loopgate Harness (py_ralph_frame) is an open-source Python verification harness and orchestrator developed by rxdt. It is built to enforce deterministic quality gates and testing boundaries around both headless CLI coding agents and interactive development agents working within a codebase. In daily use, the harness integrates into local Git workflows via dedicated Git hooks and terminal commands such as 'harness init' and 'configure-agents'. When an agent generates code, Loopgate intercepts and validates the changes by evaluating staged diff sizes and executing configured validation pipelines, including Ruff formatting, Ruff linting, Hypothesis property-based tests, and Mutmut mutation testing. Loopgate differentiates itself from standard agent execution runners by focusing specifically on strict quality enforcement and mutation scoring. It prevents automated agents from introducing silent regressions or out-of-scope edits by maintaining a strict configuration hierarchy across standalone config files, pyproject.toml, tox.ini, and setup.cfg. Loopgate is free and open-source software built for Python 3.10 through 3.14 environments and is currently under active early-stage development.
Dual Agent Mode Support — Applies automated validation rules across both headless CLI agent runs and interactive human-agent sessions. - Git Hook Gating — Configures local Git hooks in .githooks to intercept and validate staged changes before commits are finalized. - Mutation Score Tracking — Integrates Mutmut mutation testing into the validation gate to track mutation scores and catch regressions. - Property-Based Testing — Integrates Hypothesis test execution to verify agent-generated code against edge cases. - Staged Diff Guard — Constrains agent modifications by monitoring and bounding the diff size of staged files. - Config Precedence Resolution — Manages validation tool settings through a hierarchy honoring standalone configs, pyproject.toml, tox.ini, and setup.cfg.
Use cases
Use Case: Gating Autonomous Agent Commits A developer runs a headless coding agent to resolve repository issues. Loopgate's Git hooks inspect the staged changes, ensure diff sizes remain within safe thresholds, and reject the commit if linters or test gates fail.
Use Case: Enforcing Mutation Testing on Agent Changes A team wants to verify that agent-written code contains meaningful tests rather than superficial coverage. Loopgate executes Mutmut to evaluate mutation scores and gates PRs or commits based on test mutation resilience.
Use Case: Standardizing Python Linter and Formatter Rules A project maintainer configures Loopgate via pyproject.toml to enforce ruff_lint and ruff_format rules consistently whenever an AI agent attempts to modify Python source files.
Who it is for
Python software engineers, maintainers, and researchers orchestrating autonomous or interactive AI coding agents who need automated guardrails, mutation testing, and diff boundaries. It is not intended for developers seeking a full-featured IDE or non-Python agent environments.
Install guide
Installation Install the harness into your Python virtual environment (Python 3.10 to 3.14): bash git clone https://github.com/rxdt/loopgate_harness.git cd loopgate_harness pip install -e .
First Run Initialize Loopgate inside your target repository to configure Git hooks and generate the configuration scaffold: bash # Initialize the harness and Git hooks harness init # Configure agent rules and environment injection harness configure-agents
Example Session Run the gating checks manually against staged changes or verify hook configurations: bash # Stage your changes or let the agent stage changes git add . # Run the verification gate manually harness gate # Git commit will automatically trigger the gate via .githooks git commit -m "feat: implement new endpoint with agent"
SWE-agent — SWE-agent is an open-source autonomous software engineering agent harness built and maintained by researchers from Princeton University and Stanford University.…
Weaver — Weaver is an open-source agent harness and orchestration framework maintained by sean35mm on GitHub. Built on top of the Bun…
Conductor — Conductor is a multi-agent orchestration harness developed by Melty Labs. It is designed to manage, coordinate, and execute parallel coding…
Hivelore — Hivelore is an open-source agent harness and contextual memory orchestrator created by Doucs91. It provides persistent project memory, decision tracking,…
Ona — Ona (formerly Gitpod's agentic platform, now part of OpenAI) is an enterprise-grade cloud agent orchestration platform that manages and runs…
Context Bridge — Context Bridge is an open-source CLI agent harness and context orchestration utility developed by serdardb. It is built to facilitate…