Track issues, plan work, and manage context across AI coding sessions. Local-first CLI and MCP server that keeps your agents productive.
Create and manage issues with priorities, labels, and dependencies. AI agents can claim, update, and complete tasks autonomously.
Define project plans with success criteria. Link issues to plans and track progress toward your goals.
Track AI coding sessions with automatic context preservation. Resume conversations exactly where you left off.
Create named checkpoints of your work state. Restore to any checkpoint instantly if something goes wrong.
Save decisions, reminders, and notes that persist across sessions. Never explain the same thing twice to your AI.
Store commands, configs, and notes that all agents can access. Build up institutional knowledge for your codebase.
Find context by meaning, not just keywords. Search across all sessions.
Organize context items with tags. Filter checkpoints by work stream.
Define blocking relationships. Auto-unblock when dependencies resolve.
Agents claim and release issues. Coordinate work without conflicts.
Relevance-ranked context selection with scoring, diversity, and token-budget packing.
Prepare for long conversations. Preserve critical context automatically.
One command to install. Your data stays on your machine.
Standalone binary. Use from any terminal or shell script.
sc52 tools for Claude, Cursor, Windsurf, and other MCP clients.