菜单
Manage work in the local Tickets filesystem tracker. Use when the user asks to inspect, search, create, claim, update, move, rename, complete, or lint a Tickets project or ticket.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Coordinate parallel ticket workers as an event-driven control plane through repos worktrees, Pi sessions, delegated pull-request review, squash merges, and filesystem tracker frontiers. Use when orchestrating implementation tickets rather than implementing or reviewing one ticket directly.
Implement one ticket handed off by the orchestrator and open a PR for review.
Plan a huge chunk of work — more than one agent session can hold — as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.
Provides comprehensive code review guidance. Helps catch bugs, improve code quality, and give constructive feedback. Use when: reviewing pull requests, conducting PR reviews, code review, reviewing code changes, establishing review standards, mentoring developers, architecture reviews, security audits, checking code quality, finding bugs, giving feedback on code.
Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.
Use Datadog's MCP server through mcp2cli for agent-native Datadog investigation workflows: DDSQL, Kubernetes resource inspection, metric context/tag discovery, span/log/RUM attribute discovery and aggregation, monitor authoring/validation, dashboard widgets, notebooks, incidents, services, dependencies, and error tracking. Prefer pup-datadog-cli for stable scriptable Datadog API queries.
| name | tickets |
| description | Manage work in the local Tickets filesystem tracker. Use when the user asks to inspect, search, create, claim, update, move, rename, complete, or lint a Tickets project or ticket. |
Use the Tickets CLI and its local filesystem workspace to manage ticket work predictably.
Use the Tickets CLI to discover the selected project, then use tickets list, tickets search, and tickets show to inspect the relevant work. The first line from tickets show is the resolved ticket's absolute path, followed by its complete document. Consult tickets --help or command-specific help when you need command syntax. Do not proceed until you know the exact project and tickets you will operate on.
Use CLI commands for creation, rename, movement, completion, query, and lint. Edit Markdown directly for ticket bodies, assignment, tags, parent, and blockers. Never reimplement an existing CLI operation with ad hoc shell commands or YAML manipulation.
After directly editing standard metadata, run tickets lint; the edit is complete only when lint passes. Body-only edits do not require lint.
Read the ticket and confirm Assigned-To is empty before claiming it. Never overwrite another assignee. Use the human or agent name supplied by the user, project, or harness and preserve it exactly. If none is supplied, choose one recognizable name and reuse it throughout that session. Write it directly to Assigned-To, then run tickets lint.
Move the ticket to another status only when the user or project conventions identify the work status.
When the ticket's requested work is complete, use tickets done. Completion is reached only when the command succeeds and its cleanup finishes.
