# Loop Retro

> Session retrospective that mines the current session — the main transcript plus every subagent transcript — for concrete ways to improve the kit capabilities (skills/agents) that ran. Spawns a retro agent to read the large transcripts off the resident's context and write its findings to a per-run file: per capability, what worked and where its instructions caused friction, each tied to transcript evidence. Edits no code or capability. Auto-runs at the end of a /full-loop-* flow, or invoke directly after a session that exercised kit skills/agents.

- Skill: `superliaye/loop-retro` (Agent Skill)
- Install (CLI): `npx get-skill add superliaye/loop-retro`
- Raw SKILL.md: https://api.skillmd.com/api/skills/superliaye/loop-retro/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: superliaye (https://skillmd.com/u/superliaye)
- Updated: 2026-11-21
- Page: https://skillmd.com/skills/superliaye/loop-retro

---


# /loop-retro

Turn a finished session into improvement opportunities for the skills and agents it used.
The transcripts are large, so you (the **resident agent**) hand the reading to a **retro
agent**, which writes its findings to a file and hands you back the path — you spend
almost no context yourself.

## Spawn the retro agent

Give the agent the working directory and, when the Codex runtime exposes it, the resident's root
thread id:

```
Agent({
  subagent_type: "loop-retro-agent",
  description: "retro on this session",
  prompt: `
    WORKING DIRECTORY: <cwd> — the repo this session ran in.
    CODEX ROOT THREAD ID: <root id when available; otherwise unavailable>
  `
})
```

In a root Codex session, `$CODEX_THREAD_ID` is the root id. If no root id is available, the child
derives it from its own Codex thread metadata; do not guess from working directory or timestamps.

## Point the user at the findings

The retro agent writes its findings to a file and returns the path. Print that path and
its one-line headline — **don't paste the findings into chat**. Apply nothing
automatically; the human opens the file and decides what to act on. If the agent found no
readable capability body in the session, relay that plainly.

