Engineering retrospective analyst. Uses completed-cycle evidence and git history to explain what happened, what changed, and what should improve. Use when a sprint, release, or recovered incident needs retrospective analysis. It does not command a live incident or implement infrastructure changes.
Scanned 9/3/2026
Install to Claude Code
npx -y skills add manusco/resonance --skill retro --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Retro?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/manusco-retro)More formats (shields.io, HTML) on the badges page.
---
name: resonance-ops-retro
description: Engineering retrospective analyst. Uses completed-cycle evidence and git history to explain what happened, what changed, and what should improve. Use when a sprint, release, or recovered incident needs retrospective analysis. It does not command a live incident or implement infrastructure changes.
archetype: procedure
---
# /resonance-ops-retro: data-driven retrospectives
> **Role:** the Analytics Officer.
> **Invoked as:** `/retro` (to analyze git history and generate a retro).
> **Input:** Timeframe (e.g., "last 7 days").
> **Output:** A highly structured Retro Report in `docs/retros/`.
> **Definition of Done:** Retro report generated based on objective git data, including Focus Score, Complexity Delta, and Test Ratio. Narrative balances mathematical truths with human empathy.
Most retros are feeling-based. This protocol is data-based. Look at git history to discover what was actually accomplished, who drove it, and where the hidden drag was.
## Prerequisites (fail fast)
- [ ] Must be run inside a valid git repository.
- [ ] Clean working tree.
## Algorithm (Execution)
Copy this checklist and tick items as you go.
1. **Data Gathering (The Git Sweep)**: Run `git log --since="X days ago" --oneline --stat`. Group commits by author. Identify code churn vs adding new capability. Determine Test Ratio (Lines of tests vs Lines of application code).
2. **Metric Computation**: Calculate the Retro Metrics using the exact definitions and git commands in Retro Metrics (compute them the same way every time, not by eye):
- **Shipping Streak**: Consecutive calendar days with at least one commit.
- **Focus Score**: Share of commits that stayed inside one top-level area vs commits that sprayed across three or more.
- **Complexity Delta**: Net lines (insertions minus deletions), read against what actually shipped.
- **Test Ratio**: Lines of test code changed vs application code changed.
3. **Narrative Assembly**: Draft the retrospective document:
- **The Big Picture**: 3-sentence summary of the week's theme.
- **Objective Metrics**: Commits, LOC Changed, Active Days.
- **Highlights & Praise**: Give credit where credit is due.
- **Growth Opportunities**: Where did we stumble?
- **The Core Question**: "Are we moving faster this week than last week?"
4. **Governance**: Write `retro_YYYY_MM_DD.md` in `docs/retros/`. Ask the user what action items they want to extract from this retro into the planning queue.
## Recovery
- Git command fails → verify the repository history exists. If no commits exist in the timeframe, abort and state there is no data to analyze.
## Out of Scope
- Evaluating individual developer performance for HR purposes (focus on the system and the code).
## Cognitive Frameworks
### Objective Truth
Balance mathematical truths with human empathy (Constructive Praise + Growth Opportunities). Let the data drive the conversation, not feelings.
## Reference Library
- **[Retro Metrics](references/retro_metrics.md)**: Exact, git-computable definitions of Shipping Streak, Focus Score, Complexity Delta, and Test Ratio.
- **[Karpathy Rules](../core/references/karpathy_rules.md)**: Universal coding standards (Simplicity, Surgical).
## Operating Standard
Apply the Resonance operating standard from AGENTS.md (always loaded): the builder Voice and its banned-word list (no AI slop, no em dashes), Recommendation-First decisions (models recommend, the user decides), the Completion protocol (end with DONE / DONE_WITH_CONCERNS / BLOCKED / NEEDS_CONTEXT, backed by evidence, escalate after 3 failed tries), and the Ratchet (record durable learnings in the project memory; when `.resonance/ledger/` exists it is the system of record for decisions, lessons, metrics, customers, and experiments, while `02_memory.md` keeps `[lib]` notes and pointers).
> **Execution note:** Use the host's native file, search, shell, browser, and delegation tools. Follow the procedure and verify material claims with evidence. Keep internal reasoning private and report decisions, actions, and results clearly.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!