hermes-labs-ai/hermeneutic

8 stars · Last commit 2026-09-20

Mines corrections from your AI agent's chat logs to catch recurring drift and overclaiming before the next response ships. A cheap-to-expensive pre-flight gate — regex, then structured scoring, then a pressure probe — plus memory that carries past corrections into similar tasks. MIT, zero dependencies.

README preview

# hermeneutic

**Catch recurring AI drift using corrections already present in your chat logs.**

[![PyPI](https://img.shields.io/pypi/v/hermeneutic.svg)](https://pypi.org/project/hermeneutic/)
[![Python](https://img.shields.io/badge/python-%E2%89%A53.10-3776AB)](https://www.python.org/)
[![CI](https://github.com/hermes-labs-ai/hermeneutic/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/hermes-labs-ai/hermeneutic/actions/workflows/ci.yml)
[![License: Apache-2.0](https://img.shields.io/badge/license-Apache--2.0-2ea44f)](https://github.com/hermes-labs-ai/hermeneutic/blob/main/LICENSE)

**Product page:** [hermes-labs.ai/hermeneutic](https://hermes-labs.ai/hermeneutic)

## Install

```bash
pip install hermeneutic==0.1.12
```

```bash
printf '%s\n' 'Done — shipped 14 files, all tests pass.' | hermeneutic gate
```

View full repository on GitHub →