Hedde/trigger_tree

6 stars · Last commit 2026-08-02

Documentation-discovery telemetry for Claude Code — heat/cold maps, health grade, evidence-backed router fixes. 100% local, zero tokens. /tt

README preview

# 🌳 trigger-tree

<p align="center"><img src="https://raw.githubusercontent.com/Hedde/trigger_tree/main/assets/trigger-tree-logo.png" alt="trigger-tree logo" width="160"></p>

> **Gate documentation discovery in CI. Measure whether your project instructions change what agents do.**

[![CI](https://github.com/Hedde/trigger_tree/actions/workflows/ci.yml/badge.svg)](https://github.com/Hedde/trigger_tree/actions/workflows/ci.yml) [![coverage](https://img.shields.io/endpoint?url=https%3A%2F%2Fraw.githubusercontent.com%2FHedde%2Ftrigger_tree%2Fbadges%2Fcoverage.json)](https://github.com/Hedde/trigger_tree/actions/workflows/ci.yml) [![release](https://img.shields.io/github/v/release/Hedde/trigger_tree?label=release)](https://github.com/Hedde/trigger_tree/releases/latest) [![PyPI](https://img.shields.io/pypi/v/trigger-tree?label=pypi)](https://pypi.org/project/trigger-tree/) [![platforms](https://img.shields.io/badge/platforms-macOS%20·%20Linux%20·%20Windows-lightgrey.svg)](docs/platform-support.md) [![docs discoverability](https://img.shields.io/endpoint?url=https%3A%2F%2Fraw.githubusercontent.com%2FHedde%2Ftrigger_tree%2Fbadges%2Fdocs-discoverability.json)](docs/ci-gate.md) [![docs health](https://img.shields.io/endpoint?url=https%3A%2F%2Fraw.githubusercontent.com%2FHedde%2Ftrigger_tree%2Fbadges%2Fdocs-health.json)](docs/heat-model.md)

Two concrete offers:

1. **CI gate — immediate.** A deterministic, telemetry-free structural check fails a
   PR with the affected file and a suggested fix.
2. **Instruction adherence — after a few working sessions.** See whether an applicable
   `CLAUDE.md` directive was followed by observable behavior, and what that always-loaded
   instruction costs on every request.

Heat maps and the A–F documentation-health grade add longer-term discovery evidence.
Everything runs locally with Python’s standard library: no cloud, analytics, network
calls, or model tokens while measuring.

View full repository on GitHub →