Skip to content
ido4 ido4

Retrospective Skills

ido4 provides methodology-specific retrospective skills that analyze completed execution containers using real governance data.

SkillMethodologyWhat It Analyzes
/ido4dev:retroHydro (alias)Completed wave
/ido4dev:retro-waveHydroWave delivery, epic integrity, velocity
/ido4dev:retro-sprintScrumSprint delivery, type distribution, DoR/DoD compliance
/ido4dev:retro-cycleShape UpCycle delivery, bet outcomes, circuit breaker analysis

How Retros Work

All retrospective skills follow the same pattern:

  1. Call composite data tools to gather governance data
  2. Compute real metrics from the audit trail (not estimates)
  3. Analyze patterns — what went well, what didn’t, what to change
  4. Persist findings to memory for future reference
  5. Present a structured retrospective report

Every claim is backed by data. If the retro says “velocity improved by 15%”, it’s because the audit trail shows 15% more tasks completed per unit time.

/ido4dev:retro-wave (Hydro)

/ido4dev:retro
/ido4dev:retro wave-001-foundation

Defaults to the most recently completed wave. Analyzes:

  • Delivery: Real throughput (tasks/day), cycle time distribution
  • Epic Integrity: Were epics kept cohesive? Any split-epic incidents?
  • Blocking Time: Measured duration tasks spent blocked, root cause analysis
  • Actor Patterns: Who did what (agent-alpha did 60% of transitions)
  • Governance Quality: Compliance score for the wave period
  • Dependency Health: Blocking chains, satisfaction timing
  • Velocity Trend: Compared to previous waves

Output

RETROSPECTIVE -- wave-001-foundation (COMPLETED)


DELIVERY:
  Throughput: 1.4 tasks/day (5 tasks in 3.5 days)
  Cycle time: mean 2.3d, median 1.8d, p95 4.1d


OUTLIER: #3 (Config management) took 4.1 days -- dependency on
  #1 wasn't declared, causing a 2-day delay mid-implementation


BLOCKING: 1.2 days total blocking time
  #4 blocked 0.8 days (waiting on external API access)


ACTOR ANALYSIS:
  agent-alpha: 9 transitions (60%) -- primary executor
  agent-beta: 4 transitions (27%) -- review-focused
  human:bogdan: 2 transitions (13%) -- approvals only


GOVERNANCE: 95/A for wave-001 period
  All transitions followed proper workflow
  Zero overrides used


RECOMMENDATIONS:
  1. Declare dependencies during refinement
  2. Establish SLA for external access requests

Key Principle

Every recommendation is data-backed with a specific action:

  • Not: “We should communicate better”
  • Instead: “Task #38 was blocked for 4.2 days waiting on API spec from #35. Enforce refinement step requiring API contracts before tasks enter Ready for Dev.”

/ido4dev:retro-sprint (Scrum)

/ido4dev:retro-sprint
/ido4dev:retro-sprint "Sprint 14"

Scrum-specific analysis:

  • Sprint Goal Achievement: Was the sprint goal met?
  • Spillover: Items carried over from previous sprint — how much, why
  • Type Distribution: Story vs Bug vs Spike vs Tech Debt ratio
  • DoR Compliance: Items entering sprint without full Definition of Ready
  • DoD Compliance: Items approved without full Definition of Done (per type)
  • Velocity: Story points delivered vs committed
  • Pipeline Effectiveness: Did type-scoped pipelines catch quality issues?

/ido4dev:retro-cycle (Shape Up)

/ido4dev:retro-cycle
/ido4dev:retro-cycle cycle-001-notifications

Shape Up-specific analysis:

  • Bet Outcomes: Which bets shipped, which were killed, why
  • Circuit Breaker: Was it triggered? Should it have been triggered earlier?
  • Appetite Accuracy: Did bets fit within their appetite? Scope creep detected?
  • Scope Changes: Tasks added mid-cycle outside original pitch
  • Hill Chart Review: Where bets stalled (uphill unknowns vs downhill execution)
  • Cooldown Assessment: What’s shaped and ready for the next betting table

Memory Persistence

All retro skills persist findings to memory in structured blocks:

--- FINDINGS ---
wave-001: throughput=1.4/day, cycle=2.3d mean, compliance=95/A
pattern: undeclared dependency caused 2-day delay
baseline: 1.4 tasks/day is the velocity reference for wave-002
--- /FINDINGS ---

These findings create a velocity baseline that future planning and standup skills reference for capacity estimation and trend detection.