Runs a structured retro that produces actionable items assigned to specific roles.
From sdlc-cross-rolenpx claudepluginhub sethdford/claude-skills --plugin sdlc-cross-roleThis skill uses the workspace's default tool permissions.
Designs and optimizes AI agent action spaces, tool definitions, observation formats, error recovery, and context for higher task completion rates.
Enables AI agents to execute x402 payments with per-task budgets, spending controls, and non-custodial wallets via MCP tools. Use when agents pay for APIs, services, or other agents.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
Runs a structured retro that produces actionable items assigned to specific roles.
Retrospectives are where teams reflect on how they work and commit to changes. But many retros become venting sessions with no follow-up. The difference is structure: retrospective-to-action ensures that:
ISO/IEC 12207 Reference:
Output: Retro framing
Ask the team to capture observations in three categories:
What Went Well?
What Didn't Go Well?
What Confused Us?
Output: Observation lists from all team members
For each "didn't go well" observation:
Output: Root causes and potential improvements
For each root cause:
Output: Improvement ideas with owners and effort estimates
Quick wins (< 1 hour) — Do this immediately
Sprint work (1-5 days) — Add to next sprint backlog
Epic work (> 5 days) — Plan for future sprint
High impact — Affects multiple people, prevents recurrence of major issue
Medium impact — Affects some people, improves efficiency
Low impact — Nice to have, low urgency
Output: Prioritized improvement list
For each improvement:
Output: Assignments
Retrospective: Sprint [N]
Date: [Date]
Team: [Members]
Sprint Goals: [Goals]
Goal Achievement: [Yes / Partial / No]
Key Events: [Incidents, blockers, changes]
What Went Well:
• [Observation] - Team sentiment: [positive]
• [Observation] - Team sentiment: [positive]
• [Observation] - Team sentiment: [positive]
What Didn't Go Well:
• [Observation] - Root cause: [cause] - Improvement: [idea]
• [Observation] - Root cause: [cause] - Improvement: [idea]
• [Observation] - Root cause: [cause] - Improvement: [idea]
What Confused Us:
• [Observation] - Clarity improvement: [idea]
• [Observation] - Clarity improvement: [idea]
Improvements (Prioritized):
Quick Wins (< 1 hour):
[ ] [Improvement] - Owner: [name] - Effort: [estimate]
[ ] [Improvement] - Owner: [name] - Effort: [estimate]
Sprint Work (1-5 days):
[ ] [Improvement] - Owner: [name] - Effort: [estimate]
[ ] [Improvement] - Owner: [name] - Effort: [estimate]
Epic Work (> 5 days):
[ ] [Improvement] - Owner: [name] - Effort: [estimate]
Backlog Items Created:
[ ] [Issue ID] - [Title] - [Owner] - [Sprint]
[ ] [Issue ID] - [Title] - [Owner] - [Sprint]
[ ] [Issue ID] - [Title] - [Owner] - [Sprint]
Follow-Up:
Next retro date: [Date]
Review improvement status: [Yes / No]
Retros without follow-up
Retros that blame individuals
Treating all observations as equally important
Improvements with no owners
Not reviewing improvement status in the next retro
Retros that don't inform process changes