Claude-native starter kit that bootstraps multi-agent team infrastructure for any project. Clone once, install as a global skill, run /kit-install in any project directory. Includes: - 3-tier profile system (minimal/standard/full: 3-12 agents) - 16 agent archetype templates with personality spectrum - 18 skill templates using domain-action naming convention - Stakeholder persona panel for workshops and PR reviews - SQLite ticketing DB with CLI tools (config-based DB paths) - Decision tracking, sprint lifecycle, workshop orchestration - Multi-git-host support (GitHub, Gitea, GitLab) - /kit-update skill for syncing with source repo evolution - Naming theme support for agent identity/flavor - Smoke tests for all three profile tiers Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2.3 KiB
2.3 KiB
name, description, user-invocable, allowed-tools
| name | description | user-invocable | allowed-tools |
|---|---|---|---|
| sprint-retro | Run a sprint retrospective. Use when the user says "run retro", "retrospective", "what went well", or invokes /sprint-retro. Scans git history and ticket outcomes for the sprint period, generates a structured retro document. | true | Bash, Read, Grep, Glob, Write, Task |
Sprint Retrospective
Run a structured sprint retrospective by gathering data from git history, ticket outcomes, and PR activity.
Workflow
1. Identify Sprint
Find the most recently completed sprint, or accept --sprint N as argument.
- Read sprint record from DB:
db/connectors/sprint status - If a sprint number is provided, use:
db/connectors/ticket list --sprint N - Extract sprint start and end dates from the sprint record.
2. Gather Data
Collect activity data for the sprint period:
Git history:
git log --oneline --since=<start_date> --until=<end_date>
Ticket outcomes:
db/connectors/ticket list --sprint N
Categorize tickets into: done, carry-over, cancelled.
PR history:
- Count merged PRs in the period
- Calculate average review turnaround (time from open to merge)
Decision changes:
git log --oneline --since=<start_date> --until=<end_date> -- decisions/
3. Generate Retro Document
Write the retrospective to docs/sprints/sprint-N/retro.md using this structure:
# Sprint N: <Theme> — Retrospective
**Period:** <start_date> to <end_date>
**Velocity:** X/Y tickets completed (Z%)
## What Shipped
- List of completed tickets with brief descriptions
## What Went Well
- Smooth integrations, clean PRs, good velocity areas
## What Hurt
- Blockers, stalled tickets, context switching, scope creep
## What We Learned
- Process insights, technical lessons, team dynamics
## Carry-Over
| # | Title | Why it carried |
|---|-------|----------------|
## Action Items
- Concrete improvements for next sprint
4. Optional: Stakeholder Perspective
If stakeholder personas are configured, spawn 1–2 personas to comment on what shipped from a user perspective. Use the Task tool for parallel persona evaluation.
5. Report
Write the retro file to disk and report a summary to the user including:
- Velocity percentage
- Key highlights from each section
- Number of action items generated