retro

A structured review held after a release, milestone, bug, or incident to record what worked, what did not, and what to change next. The results are saved in a retrospective document.

In plain words
What is it for?
Use it after shipping or at the end of a sprint to compare planned and completed work, inspect changes, and create action items.
Why use it?
It captures lessons while events are still fresh and turns them into specific follow-up actions instead of letting them be forgotten.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/dwarvesf/dwarves-kit/retro
Clone the repo
git clone --depth 1 https://github.com/dwarvesf/dwarves-kit
Per session 32 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,772 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00032 $0.01772
Opus 5 $0.00016 $0.00886
Sonnet 5 $0.00006 $0.00354
Haiku 4.5 $0.00003 $0.00177

Measured 2d ago against content hash c037b28bcc74, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

retro scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/retro.md · 153 lines

How it starts

The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a retrospective facilitator. The team just shipped something. Your job is to extract learnings before context is lost.

This is NOT a celebration. This is NOT a blame session. This is a structured extraction of signal.

When to run

  • After /kit:ship completes
  • At the end of a sprint or milestone
  • After a significant bug or incident
  • Periodically (weekly or biweekly) for ongoing projects

Process

Bracket the phase for timing (SPEC-129) before starting: bash lib/gate/gate-ledger.sh outcome <rid> Reflect start.

Step 1: Gather data

Collect facts before opinions:

# Recent commits
git log --oneline --since="1 week ago"

# Files changed
git diff --stat main

# Spec completion
cat docs/specs/SPEC-NNN-<slug>.md  # check task completion rate

Present a summary:

  • Tasks planned vs completed
  • Files added/changed/deleted
  • Commits count and pattern (are they atomic or messy?)
  • Time span (when did work start vs ship)

Step 1b: Doc-impact + completeness sweep

Run the pinned diff (the integration branch's merge-base) against the WORKFLOW doc-impact map and list any companion doc the diff should have updated but did not. Also read ~/.claude/dwarves-kit/logs/completeness.log and surface un-cleared decision/doc warnings from this cycle. Report the gaps as retro signal, not a block; a recurring gap is a candidate to promote the clause to a hook (PHILOSOPHY section 5 bar). Source: SPEC-006.

Step 1c: Decision-capture nudge (advisory)

Ask once: did this cycle make a non-obvious, reversible-with-cost decision that is NOT already recorded in a SPEC Decision Log or an existing ADR? Examples: a storage shape, a library choice, an error contract, an architectural boundary, a deliberate scope cut. A decision already in a spec's ## Decision Log or an existing docs/decisions/NNNN-*.md does not need a second home; this is only for the ones that fell through.

If yes, suggest drafting docs/decisions/NNNN-<slug>.md (next number after the highest existing; match the style of the existing ADRs in docs/decisions/, there is no separate template file) capturing: context, the decision, why, and the rejected alternatives in one or two lines each.

Read the full file on GitHub · 153 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 153 lines · 32 tokens per session scan A c037b28bcc74

Subscribe to this mod's changes

retro is a command published in the GitHub repository dwarvesf/dwarves-kit (11 stars, last pushed 2d ago), licensed MIT. It adds 32 tokens to every session and 1,772 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.