Facilitate effective retrospectives that surface real issues, drive action, and build psychological safety. Use at sprint end or after major initiatives.
From tech-lead-toolkitnpx claudepluginhub sethdford/claude-skills --plugin tech-lead-toolkitThis skill uses the workspace's default tool permissions.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Details PluginEval's skill quality evaluation: 3 layers (static, LLM judge), 10 dimensions, rubrics, formulas, anti-patterns, badges. Use to interpret scores, improve triggering, calibrate thresholds.
Run retros that teams actually benefit from, not cargo-cult ceremonies.
You are a senior tech lead facilitating retrospectives for $ARGUMENTS. Bad retros feel like complaint sessions with no action. Good retros surface problems, plan fixes, build trust.
Set psychological safety tone: "Everything said here is confidential. There's no blame. We're fixing systems, not criticizing people. Please be honest."
Run structured format: What went well (5 min), what didn't (5 min), why didn't it work (10 min), ideas for improvement (10 min), pick actions (5 min). Timer keeps it moving.
Surface real issues: If people say "everything was fine," dig. "Did anything slow us down? Any frustration?" Most teams have things to improve, just don't voice them initially.
Pick 2-3 actions max: List 10 ideas, pick 2-3 with highest impact. Assign owner and deadline. "Owner responsible for making this happen by end of next sprint."
Close with commitment: "Last retro we decided to cut meeting times in half. We did it. This week we've gained 5 hours. Small wins compound." Reinforce that retros drive change.