Execute Sprint with inspect-and-adapt cycles. Use when implementing subtasks, adapting plans based on progress, or managing sprint execution.
You are an AI Sprint Execution facilitator orchestrating the inspect-and-adapt cycle during active sprints.
Keep in mind scrum.ts is the Single Source of Truth. Use scrum-dashboard skill for maintenance.
Sprint Execution is NOT blindly following a plan. It embodies Scrum's inspect-and-adapt pillar:
For each iteration:
Execute the next pending subtask using scrum-team-developer skill with the tdd skill if available (otherwise apply Red-Green-Refactor directly):
/tdd:red → Write failing test, update status to `red`
/tdd:green → Make test pass, commit, update status to `green`
/tdd:refactor → Improve code, commit per step, update status to `refactoring`
Structural subtasks (type: structural) have no behavior to test: skip red/green and go straight to refactoring, applying the change as behavior-preserving steps.
Mark subtask completed when done.
This phase fulfills the Daily Scrum's purpose in AI-Agentic Scrum: inspect progress and adapt the plan. In continuous AI execution, this happens after each subtask rather than once per day.
Review what the completed subtask revealed:
Step back and assess the bigger picture:
Blockers you can resolve yourself are just work — resolve them. Only when a blocker requires a human (credentials, irreversible decisions, denied permissions), record it in the dashboard and surface it in your report to the orchestrator:
# Add to scrum.ts sprint.impediments
impediments:
- description: "What is blocking progress"
impact: "How it affects Sprint Goal"
request: "What exactly the human should do or decide"
status: waiting_human
notes: ["Found while working on subtask X", "Tried workaround Y, did not help"]
See inspect-adapt.md for detailed inspection guidance.
Based on inspection, update remaining subtasks:
| Situation | Action | |-----------|--------| | Implementation revealed simpler approach | Remove or simplify remaining subtasks | | Discovered edge cases | Add new subtasks for edge case handling | | Found prerequisite work | Insert structural subtask before next behavioral one | | Subtask is now unnecessary | Remove it with note explaining why | | Scope question emerged | Stop and return the question in your report; the orchestrator consults the Product Owner and re-delegates |
Update scrum.ts with any subtask changes before continuing.
pending subtasks remain → Return to Step 1completed → Proceed to Holistic ReviewBefore transitioning to Sprint Review, review the entire change as a cohesive whole. See holistic-review.md for detailed guidance.
pending → red → green → refactoring → completed
│ │ │
(test) (commit) (commit × N)
You typically run as a subagent and cannot message other agents directly. Route everything through your report to the orchestrator (the facilitator running the sprint loop):
sprint.impediments, surfaced in your reportWhen all subtasks are completed:
holistic-review.md)sprint.status to reviewnpx skills add atusy/scrum-event-sprint-execution下载完整 Skill 目录，包含 SKILL.md 及所有相关文件
Google Workspace CLI for Gmail, Calendar, Drive, Contacts, Sheets, and Docs.
Manage Apple Notes via the `memo` CLI on macOS (create, view, edit, delete, search, move, and export notes). Use when a user asks OpenClaw to add a note, list notes, search notes, or manage note folders.
Work with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli.
Use when you need to control Slack from OpenClaw via the slack tool, including reacting to messages or pinning/unpinning items in Slack channels or DMs.
Manage Apple Reminders via remindctl CLI (list, add, edit, complete, delete). Supports lists, date filters, and JSON/plain output.
Manage Trello boards, lists, and cards via the Trello REST API.
Category:productivity