You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The open-source governance layer for AI agents. No model in the coordination loop: a plan replays to the same task graph, in per-task git worktrees; offline-verifiable run receipts, signed lineage + an opt-in HMAC audit chain. Claude Code, Codex, Gemini CLI, +40 more. Any deliverable, code to audit evidence. Beta. https://bernstein.run
Autonomous multi-change orchestration for Claude Code — spec in, merged features out. Sentinel supervisor, parallel worktrees, developer memory, and more.
Local coding-agent orchestrator — DAG of auto-approved, git-worktree-isolated sub-sessions across LLM providers (Claude/Kimi/Grok/DeepSeek/local). AGPL-3.0.
Production-tested agent skills for Claude Code, Codex, Cursor, and OpenCode — session management, multi-session reconciliation, worktree recovery, PR review, and skill scaffolding. The proportionate-review leg of trust infrastructure for agents.