TICK.md is a Markdown-first task coordination system backed by Git. It provides a CLI and an MCP server that lets humans and AI agents create, claim, and complete tasks while keeping an automatic git-backed audit trail. It includes file-locking to prevent edit collisions, dependency tracking, and visualization tools for task graphs.
Use TICK.md when multiple agents (human or bot) need to coordinate work on the same repo or project. It's suited for engineering teams, distributed bot fleets, or any setup where traceable, claimable tasks and reproducible change history matter. Trigger on workflows requiring guarantees against concurrent edits or when you need an explicit claim/comment workflow for agents.
Designed for OpenClaw/Claude-style agents that can call CLI tools or run an MCP server; integrates with any agent able to execute the tick CLI or speak to the MCP server.
TICK.md is a Git-backed Markdown task coordination CLI and MCP server for multi-agent workflows. It provides task claiming, dependency tracking, file locking, and automatic git commits for audit trails. No bundled scripts were present to test. The SKILL.md is well-written with clear instructions, but it's essentially documentation for an npm package rather than a self-contained skill with automation scripts.
No security concerns found — no remote code execution patterns, no hardcoded credentials, no exfiltration. The skill is documentation-only (no bundled scripts), so security scoring is based on the SKILL.md content which is clean. Architecture is decent but the SKILL.md is very long and could benefit from moving setup details to references/. Usefulness is high: multi-agent task coordination is a real, common need in the MCP/AI agent space.