Turn the agent corrections you throw away into local eval data: regression tests, failure analysis, security signals, and handoff memory. No LLM judge, no uploads, zero dependencies.

Git-native AI code provenance: records which AI agent wrote which line, signs each attribution with ed25519, stores it in your git history. Cross-agent (Claude Code, Cursor, Copilot, Codex, Windsurf, OpenCode, Gemini).

https://forgeproof.flyingcloudtech.com

AI coding agents should work in attempts, not your working tree

Local-first AI code provenance: line-level attribution, PR risk reports, policy checks, and tamper-evident audit trails for AI-assisted software.

CodeGenLink is a Visual Studio Code extension that interacts with GitHub Copilot Chat to generate code, analyze its origin, and identify the associated license.

See where your AI code takes root — and how it grows over time.

Crocking — AI authorship detector for git repositories. Analyzes commit history for statistical signatures of undisclosed AI-generated code. Zero dependencies.

The intent layer for Git — link AI agent conversations to the commits they produced, then query why any line of code was written.