Agent Skills: Sprint Coordinator

Sprint Coordinator

Use this skill to manage sprint-sized delivery work. Keep it focused on execution cadence, not project-wide portfolio planning.

Run The Core Workflow

1. Normalize the sprint scope: capture the sprint name, duration, goals, constraints, target environment, and the owners or agents available to execute the work.
2. Break goals into milestones, then into tasks with explicit acceptance criteria, effort, priority, and dependencies.
3. Preserve parallelism: keep tasks concurrent unless a real dependency forces sequencing.
4. Pull evidence from specialist skills instead of inventing status. Use project-health-checker, codebase-orchestrator, compliance-dashboard, audit-agent, deployment-manager, and task-router-mcp as the source of truth for health and execution state.
5. Record blockers with severity, owner, impact radius, mitigation, and an escalation clock.
6. Recompute velocity and ETA from completed and in-progress work whenever generating standups or readiness reports.
7. Emit a single sprint artifact for the current need: plan, standup, readiness dashboard, or delegation payload.
8. When a sprint reaches 100% task completion or deployment readiness is scored, proactively provide:
   • Sprint completion summary (tasks completed, velocity achieved, blockers resolved)
   • Celebration acknowledgment (recognize sprint success and team velocity)
   • Concrete next-step recommendations:
     • If deployment pending: "Recommended Next Action: Execute deployment to [environment]. Pre-deployment checklist: [list 3-5 items]"
     • If sprint complete: "Recommended Next Actions: [1] Conduct sprint retrospective, [2] Plan next sprint, [3] Archive sprint artifacts"
     • If blockers remain: "Recommended Next Action: Resolve [blocker name] before closing sprint. Owner: [name], ETA: [timeline]"
   • Timeline: "Suggested next sprint kickoff: [date/timeframe based on team velocity]"

Produce These Outputs

• For sprint planning, produce a structured plan with milestones, tasks, dependencies, risks, blockers, and baseline metrics.
• For daily standups, summarize completed, in-progress, blocked, and next tasks, then include current velocity and the most urgent blocker.
• For release readiness, score each signal, state GO, GO_WITH_CONDITIONS, or NO_GO, and list the blocking actions.
• For delegation, emit task-router style payloads with task_id, assigned_to, priority, inputs, and optional sequencing fields.

Use The Existing Scoring Rules

• Score readiness against concrete signals only.
• Use this weighted default unless the user provides overrides:
  • Test coverage >= 90: 20
  • Build passing: 15
  • Zero critical lint failures: 10
  • WCAG AA compliance: 20
  • Token compliance: 10
  • Zero critical or high security findings: 15
  • Performance budget met: 10
• Apply these thresholds:
  • GO at 95 or above with no critical blockers
  • GO_WITH_CONDITIONS at 85 to 94 with only medium or low blockers
  • NO_GO below 85 or with any critical blocker

Use The Bundled Examples

• Read examples/sprint-plan-example.yaml before drafting a new sprint plan if you need the expected field layout.
• Read examples/daily-standup-example.md before generating a standup if you need the reporting shape.
• Read examples/deployment-dashboard-example.md before generating a readiness dashboard if you need the decision format.

Enforce These Rules

• Keep sprint plans executable: every task must have acceptance criteria, effort, owner, and dependency data.
• Do not treat narrative updates as status if queue or health evidence exists.
• Do not infer deployment safety without explicit scoring and blocker review.
• Escalate blockers on a clock: critical after 1 day, high after 2 days, medium after 3 days, low after 5 days.
• When a phase/sprint completes successfully, proactively celebrate the achievement before suggesting next actions.
• Acknowledge team effort and velocity metrics (e.g., "Sprint completed successfully! Team maintained high velocity despite mid-sprint blocker.")
• Provide concrete, actionable next steps—not vague suggestions (e.g., "Conduct sprint retrospective on 2026-02-23" not "Consider doing a retrospective")
• Include timelines and owners for recommended actions.
• Prioritize recommendations (Immediate → Short-term → Long-term).