Show HN: Coherence – drift detector for AI-driven repos
Coherence is a Git-native drift detector designed for AI-assisted repositories. It identifies inconsistencies in codebases edited by AI agents, ensuring that code, documentation, and tests remain aligned. The tool operates locally with deterministic checks and offers an optional LLM pass for deeper analysis.
- ▪Coherence detects broken links between code, documentation, tests, and other project artifacts after AI edits.
- ▪The tool runs locally and does not send code anywhere during its checks.
- ▪Coherence provides a detailed algorithm reference and examples for understanding its drift detection signals.
Opening excerpt (first ~120 words) tap to expand
Coherence Git-native drift detector for agent-assisted repositories. Docs site: https://fireharp.github.io/coherence/ Coherence is not an AI reviewer. It is a repo consistency harness for AI-edited codebases. Tests pass. The repo still drifts. Coherence catches the broken links between code, docs, ADRs, tests, metrics, generated files, endpoints, and evidence - especially after AI-agent edits. Coherence runs locally. Deterministic checks do not send code anywhere. The optional LLM pass is disabled by default and only runs when COHERENCE_LLM=1 or --llm is set. Algorithm reference: docs/ has a long-form page for every drift meter and check, with the algorithm, JSON output shape, signal interpretation, and a benchmark scenario example. Start there if you want to understand a firing signal.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.