Hermeneutic turns your AI corrections into context for the next task
You ask a coding agent to review a function. It starts refactoring. You steer it back to the review you requested. That exchange contains useful information about your intended scope. Hermeneutic , built by Hermes Labs, mines corrections from agent conversations and surfaces relevant guidance when similar tasks come up. In this example, a later code-review request could receive guidance about…
Hermeneutic, developed by Hermes Labs, transforms AI corrections into context for future tasks. This tool mines corrections from conversations with coding agents and surfaces relevant guidance when similar tasks arise. By analyzing supported Claude Code, Codex, and OpenAI-format logs, Hermeneutic creates a local corpus of correction records, which can be inspected using the bucket command.
This allows users to identify recurring issues such as missed constraints, wrong targets, unnecessary confirmation, or scope expansion. When new prompts are processed, Hermeneutic matches them against earlier prompts using Ollama embeddings, retrieving relevant correction categories to provide concise guidance through category templates.
Each piece of advice cites its supporting correction records. To implement Hermeneutic, users need Python 3.10 or higher and Ollama. After installing the tool, users can mine their conversation history, bucket the correction records, and compile an index for retrieval. This guidance can be used directly in workflows or installed via a Claude Code hook to provide prompt context.
To get started, users can review the provided example function and report their findings. Hermeneutic is available on GitHub, and users are encouraged to try it out, follow its development, and star the repository.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
