Manage issue lifecycle with phase gates, artifacts, and checkpoint recovery.
npx skills add https://github.com/laozhong86/gxpm --skill gxpm
Please help me install this Agent Skill. Skill: gxpm Source: https://github.com/laozhong86/gxpm/tree/main/skills/gxpm Command: npx skills add https://github.com/laozhong86/gxpm --skill gxpm
Second-generation, host-native project management runtime for AI agents that manages issues through a phase-gated lifecycle, ensures auditable evidence, and enables checkpoint recovery across worktrees.
Initialize an issue context with gxpm issue context <id> and then advance through phases with gxpm issue next <id>.
High-intent search queries and answers about installing and using this skill.
AI agent project workflows use a phase-gated lifecycle to manage issues, ensuring checkpoint recovery and auditable evidence across worktrees. This approach enforces phase transitions while preserving artifacts and maintaining local state as the source of truth.
To advance issues through a project lifecycle, initialize the context using an issue ID, then transition sequentially. Agents move through triage, planning, and delivery stages while enforcing phase-specific constraints and storing results locally.
Worktree-based collaboration isolates changes by maintaining local state as the source of truth, enabling agents to resume work across sessions. This ensures recoverable workflows and preserves auditable artifacts during complex multi-issue project execution.
Yes, configuring local state as the source of truth is required for project management auditing. It guarantees traceability by enforcing artifact writes and checkpoint recovery, ensuring all phase-gated workflow transitions are auditable and recoverable.
The best way to ensure auditable evidence during multi-issue project coordination is using issue-driven lifecycle management with phase gates. This enforces artifact writes, skill loading, and local state tracking to guarantee full workflow traceability.
You should avoid phase-gated lifecycle management for AI agent workflows when projects lack complex phases, artifacts, or multi-session recovery requirements. Without the need for auditable evidence or worktree isolation, this structured approach adds unnecessary overhead.
try-worksCreate an isolated git worktree for each RLM run with Phase 0 setup.
yusaduymazRestore paused development sessions from STATE.md and checkpoint files.
stevengonsalvezExpose local Ollama models as tools via an MCP server.
lorainwingsAutomate 8-step gate checks and atomic checkpoint persistence for autopilot phase transitions.
