Explore skills

保存、恢復、列出長 session checkpoint；記錄 branch、ticket、todo、recent activity，方便中斷或 context compression 後接續。

Use when executing iterative tasks, continuous improvement, quality assurance, process optimization, or when user asks to 'apply PDCA', 'iterate and improve', 'continuous improvement cycle', 'plan-do-check-act'.

Manage your AI team. /crew = roster, /crew init = generate shared project reference, /crew add [name] "[domain]" = new member, /crew 1-on-1 [name] = check-in, /crew [name] [task] = assign task, /crew done [name] = close session, /crew remove [name] = remove member.

Creates a new SKILL.md in the correct location within this project's .ai/ structure. Trigger when the user asks to "create a skill", "add a skill for [task]", "write a skill to [do something]", or "add a new skill".

Turn an infra ticket into a lightweight spec, implementation plan, task list, and implementation notes (with optional Confluence-ready output).

Synchronizes downstream task specs after implementation. Spawned by flow-next-work after each task completes. Do not invoke directly.

Autonomous implementation engine (the "harnish" engine) that handles PRD to task decomposition, autonomous execution, cross-session context preservation, and experience accumulation, activated by various triggers.

This skill provides persistent memory and task coordination for AI coding sessions by creating a .fireteam/ folder with mission context, task state, checkpoints, and handoff notes that survive session restarts, context limits, and crashes. Use it to set up, continue, or resume a fireteam, add it to a project, coordinate agents, or split work across sessions.

Generates or updates structured Obsidian daily notes from git commit history with diff analysis. Useful for work logs, reports, and filling in past daily entries.

(Industry standard: Sequential Agent / Agent as a Tool) Primary Use Case: Delegating a well-defined task to a worker agent, verifying its execution, and repeating if necessary. Inner/outer agent delegation pattern. Use when: work needs to be delegated from a strategic controller (Outer Loop) to a tactical executor (Inner Loop) via strategy packets, with verification and correction loops.

三機(Claude/Gemini/Codex)平行審查架構，評估擴展性、模組化、安全性後投票共識

Persistent tiered memory and reminder notebook for Claude Code. On every session start, it performs memory boundary checks, monthly verified refreshes, and twice-monthly pattern extractions, with multi-window deduplication and natural language recognition to auto-manage reminders.