Origin

The product is a git-native history layer for AI-generated code that records which agent wrote every line along with the prompt, model, tokens and cost, and provides per-prompt snapshots and line-level blame to support attribution, rollbacks, and audit trails. It sells to solo developers and to engineering teams in software companies that use AI coding agents such as Claude Code, Cursor, Codex, and Gemini CLI. It is delivered as a locally-run CLI that stores sessions in git refs and notes with an optional SaaS dashboard for session replay, budgets, and policy enforcement.

Key features

• Per-prompt working-tree snapshots
• Line-level AI blame per prompt
• Auto-detects multiple AI agents
• Session capture and replay
• Prompt, model, and diff logging
• Token and cost tracking
• Git refs and notes storage
• Policy enforcement and PR checks
• Budget tracking per agent
• Audit trails for teams
• Personal dashboard with searchable history
• Zero-config git hooks setup
• Memory injection for new sessions