ai-sprint
$
npx mdskill add arcasilesgroup/ai-engineering/ai-sprintSprint lifecycle management: plan new sprints from backlog, run data-driven retrospectives comparing planned vs shipped, and track sprint-level goals. Bridges the gap between spec-level planning and day-to-day delivery.
SKILL.md
.github/skills/ai-sprintView on GitHub ↗
---
name: ai-sprint
description: "Manages sprint lifecycle: plans a new sprint from backlog, runs data-driven retros comparing planned vs shipped, checks mid-sprint goal status, generates sprint review presentations. Works with GitHub Projects and Azure DevOps. Trigger for 'start sprint planning', 'kick off the sprint', 'lets do the retro', 'what did we deliver last sprint', 'sprint goals check', 'generate the sprint review deck'. Not for daily standup; use /ai-standup instead. Not for solo PR retro; use /ai-learn instead."
effort: mid
argument-hint: "plan|retro|goals|review [--sprint name]"
mode: agent
requires:
anyBins:
- gh
- az
bins:
- python3
model_tier: sonnet
mirror_family: copilot-skills
generated_by: ai-eng sync
canonical_source: .claude/skills/ai-sprint/SKILL.md
edit_policy: generated-do-not-edit
---
# Sprint
## Purpose
Sprint lifecycle management: plan new sprints from backlog, run data-driven retrospectives comparing planned vs shipped, and track sprint-level goals. Bridges the gap between spec-level planning and day-to-day delivery.
## Trigger
- Command: `/ai-sprint plan|retro|goals`
- Context: sprint boundary (start or end of sprint), goal tracking mid-sprint.
## Pre-conditions (MANDATORY)
1. Read `.ai-engineering/manifest.yml` — `work_items` section.
2. Determine active provider (`github` or `azure_devops`).
3. Read `.ai-engineering/reference/gather-activity-data.md` for the canonical git log, PR query, and work item commands.
4. Use provider-specific config:
- **Azure DevOps**: filter by `area_path`, auto-detect current `iteration_path`
- **GitHub**: filter by `team_label`, use milestones for sprint boundaries
5. Use all standard and custom fields the platform provides.
## Workflow
Four modes follow the sprint lifecycle:
1. `plan` — read backlog, propose sprint goals, scope items, write sprint file.
2. `goals` — mid-sprint progress check vs the planned goals.
3. `retro` — data-driven retrospective comparing planned vs shipped.
4. `review` — generate the sprint review deck (delegates to `/ai-slides`).
## Modes
### plan -- New sprint planning
1. **Review backlog** -- read open specs, GitHub Issues/Projects, and prioritized items from the backlog (GitHub Issues with priority labels or manual ranking).
2. **Assess capacity** -- count working days in sprint, factor in known absences or blockers from decision-store.
3. **Select items** -- pull highest-priority items that fit capacity. Apply RICE scores from backlog prioritization.
4. **Estimate effort** -- use size labels (XS/S/M/L/XL) from issue standard. Flag items missing size estimates.
5. **Draft sprint board** -- output planned items grouped by priority:
```markdown
## Sprint: {name} ({start} - {end})
### Goals
1. {Goal 1 -- measurable outcome}
2. {Goal 2 -- measurable outcome}
### Planned Items
| # | Priority | Size | Item | Spec |
|---|----------|------|------|------|
| 1 | p1 | M | Fix hook installation on Windows | spec-054 |
| 2 | p2 | L | Add telemetry dashboard | spec-054 |
```
6. **Store** -- save sprint plan to `.ai-engineering/sprints/{name}.md`.
### retro -- Sprint retrospective
1. **Load sprint plan** -- read `.ai-engineering/sprints/{name}.md`.
2. **Collect actuals** -- use the commands from `.ai-engineering/reference/gather-activity-data.md` to scan merged PRs, completed spec tasks, and commit history for the sprint period.
3. **Compare planned vs shipped**:
- Items completed as planned
- Items carried over (not finished)
- Side quests (unplanned work that entered the sprint)
- Items descoped or deprioritized
4. **Analyze patterns**:
- Estimation accuracy: actual effort vs estimated size
- Side quest ratio: unplanned / total items delivered
- Velocity trend: items completed vs previous sprints
5. **Document learnings** -- what went well, what to change, action items.
6. **Output** -- retrospective report appended to `.ai-engineering/sprints/{name}.md`.
### goals -- Sprint goal tracking
1. **Load active sprint** -- find current sprint from `.ai-engineering/sprints/`.
2. **Check goal progress** -- for each goal, assess completion signals (merged PRs, closed issues, spec task status).
3. **Report** -- traffic-light status per goal: green (on track), yellow (at risk), red (blocked/behind).
### review -- Sprint review presentation
Generate a branded sprint review PowerPoint deck using python-pptx. Each invocation produces a NEW script tailored to current data — never reused from a static template.
1. **Determine sprint period** — resolve date range: `--sprint YYYY-MM` (calendar month), `--iteration <name>` (query provider for dates), or default to current month.
2. **Gather data** — use commands from `.ai-engineering/reference/gather-activity-data.md` for work items and git activity. Collect quality metrics via `pytest --co -q` and `ruff check . --statistics`. Compare against thresholds in `manifest.yml`.
3. **Generate python-pptx script** — new script each time. Brand constants: `AI_BG_DARK=#0B1120`, `AI_ACCENT=#00D4AA`, `AI_PRIMARY=#1E3A5F`. Typography: `JetBrains Mono` (headings), `Inter` (body). Layout: 16:9, 13.333"×7.5".
4. **Slide structure (8-14 slides)**: Title → Sprint Overview (KPI cards) → Feature Deep-Dives (one per major spec) → Quality Metrics → Risks & Next Sprint → Q&A. Every slide requires `set_notes()` for presenter view.
5. **Execute** — write to `.ai-engineering/runtime/presentations/generate_sprint_review.py`, run it, output `.ai-engineering/runtime/presentations/sprint-review-YYYY-MM.pptx`.
**Common mistakes**: reusing old script verbatim, missing speaker notes, wrong color palette, skipping pre-conditions, hardcoding dates.
## Arguments
Modes (`plan`, `retro`, `goals`, `review`) per Modes above. Flags:
- `--sprint <name>` — sprint identifier (e.g., `2026-w12`) or month (`YYYY-MM`); defaults to current.
- `--iteration <name>` — iteration name (queries provider for dates, used with `review` mode).
## Quick Reference
```
/ai-sprint plan --sprint 2026-w12 # plan sprint for week 12
/ai-sprint retro --sprint 2026-w11 # retro on last sprint
/ai-sprint goals # check current sprint goals
/ai-sprint review --sprint 2026-03 # generate March 2026 review deck
/ai-sprint review --iteration "Sprint 12" # named iteration review deck
```
## Storage
- Sprint files: `.ai-engineering/sprints/{name}.md`
- Naming convention: `YYYY-wNN` (ISO week) or custom names
## Examples
### Example 1 — plan a new sprint
User: "kick off sprint 2026-w19 from the backlog"
```
/ai-sprint plan --sprint 2026-w19
```
Reads the backlog (GitHub Projects v2 or Azure Boards), proposes sprint goals, scopes items, writes `.ai-engineering/sprints/2026-w19.md`.
### Example 2 — retro at sprint end
User: "lets do the retro for the sprint that just ended"
```
/ai-sprint retro --sprint 2026-w18
```
Compares planned vs shipped, surfaces velocity trends, identifies blockers, writes the retro section.
## Integration
Called by: user directly. Calls: `gh project item-list`, `az boards query`, `/ai-slides` (for `review` mode). See also: `/ai-standup` (daily slice), `/ai-prose content sprint-review`, `/ai-board discover`.
$ARGUMENTS
More from arcasilesgroup/ai-engineering
- ai-adviseProactive governance advisor — checks standards, decisions, and quality trends during development. Always advisory, NEVER blocks. Three modes: `advise` (post-edit), `gate` (pre-dispatch), `drift` (on-demand decision audit). Trigger for 'governance check', 'advise on this change', 'check for drift', 'is this aligned with active decisions', 'shift-left advisory'. Not for blocking gates — use /ai-verify. Not for narrative code review — use /ai-review.
- ai-analyze-permissionsUse when Claude Code keeps asking to approve commands you have already approved, when settings.local.json has grown large, or when you want to consolidate permission grants into wildcard patterns. Trigger for 'too many permission prompts', 'clean up permissions', 'audit my settings', 'consolidate allow rules'. Claude Code only — not available in GitHub Copilot, Antigravity, or Codex.
- ai-animationDesigns motion, transitions, and micro-interactions for UI components: spring animations, gestures, easing, staggers — taste-driven detail compounding. Trigger for 'animate this', 'add transitions', 'micro-interactions for', 'gesture design', 'swipe to dismiss', 'easing for this', 'stagger the'. Not for design systems; use /ai-design instead. Not for visual art; use /ai-visual instead. Not for testing animation code; use /ai-test instead.
- ai-autopilotDelivers large multi-concern specs and backlog runs autonomously: decomposes specs into sub-specs (or normalizes work items into a backlog DAG), deep-plans with parallel agents, builds a dependency DAG, implements in waves, runs a single final quality loop with one bounded quality-remediation pass (verify+guard+review on full changeset), delivers via PR. Trigger for 'implement spec-NNN end to end', 'autopilot this', 'autonomous delivery', 'decompose and ship', 'run the backlog', 'execute these GitHub issues', 'process the sprint backlog'. Invocation is the approval gate. Not for small or single-concern tasks; use /ai-build instead. Not for ambiguous requirements; use /ai-brainstorm first.
- ai-boardOperates the project board (GitHub Projects v2 or Azure DevOps): discovers configuration after install (fields, state mappings, process templates) and syncs work-item state at lifecycle transitions. Trigger for 'set up the board', 'configure our ADO board', 'discover board fields', 'move this issue to in-review', 'update the board', 'mark as in progress', 'sync the work item state'. Two subcommands: `discover` (post-install configuration write) and `sync` (lifecycle state transitions). Auto-invoked via `sync` by /ai-brainstorm, /ai-build, and /ai-pr; fail-open. Not for backlog execution; use /ai-autopilot --backlog instead.
- ai-brainstormForces rigorous design interrogation BEFORE any code: explores approaches, surfaces ambiguity, gathers evidence, produces an approved spec that becomes the contract for /ai-plan. Trigger for 'lets add X', 'how should we handle Y', 'whats the best approach', 'I am thinking about', 'what should we build for'. Not for existing approved specs; use /ai-plan instead. Not for execution; use /ai-build instead.
- ai-branch-cleanupCleans branches safely: switches to the default branch, prunes merged and squash-merged branches, syncs to remote, sweeps stale specs, rotates `.ai-engineering/runtime/` per retention policy. Trigger for 'tidy up', 'tidy branches', 'sync to main', 'delete old branches', 'start fresh', 'rotate runtime'. Auto-invoked by /ai-pr after merge. Not for committing changes; use /ai-commit instead. Not for code-level dead-code removal; use /ai-simplify instead.
- ai-buildCanonical implementation gateway: reads approved plan.md, resolves stack from manifest, deterministic-routes each task to its adapter, dispatches the build agent in an isolated worktree, runs TDD self-validation per task, then a single final quality loop with one bounded quality-remediation pass on the full changeset before /ai-pr. Trigger for 'go', 'start building', 'execute the plan', 'implement it', 'lets do this', 'build the plan', 'resume', 'continue'. Not without an approved plan; run /ai-plan first. Not for multi-concern specs needing decomposition; use /ai-autopilot instead. Not for a single function or subcomponent; use /ai-code.
- ai-codeWrites production code that satisfies stack-context standards on the first pass: interface-first design, backward-compatibility checks, lightweight self-review. Trigger for 'implement this', 'write the code for', 'add X to Y', 'build this function', 'make this work'. Not for tests; use /ai-test instead. Not for debugging; use /ai-debug instead. Not for refactoring; use /ai-simplify instead. Not for executing an approved plan end-to-end; use /ai-build (the gateway).
- ai-commitRuns the governed commit pipeline: auto-branches from protected, stages selectively, formats and lints, scans for secrets, gates docs, composes a conventional message, pushes. Trigger for 'commit my changes', 'save my work', 'push this to remote', 'stage these files', 'ship it'. Not for opening a PR; use /ai-pr instead. Not for branch hygiene; use /ai-branch-cleanup instead.