Initialize a Trio mailbox for a Kimi Code session.
Scanned 9/5/2026
Install to Claude Code
npx -y skills add albiol2004/trio-agent-loop --skill trio-init --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Trio Init?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/albiol2004-trio-init-571bd028)More formats (shields.io, HTML) on the badges page.
---
name: trio-init
description: Initialize a Trio mailbox for a Kimi Code session.
type: prompt
whenToUse: When the user asks Kimi Code to start a new Trio loop or create its mailbox
arguments:
- goal
---
# Initialize a Kimi Code Trio mailbox
Create `loop/` unless the user explicitly names another mailbox directory.
Never overwrite a mailbox whose `STATE.md` indicates that it may be running.
Write:
- `GOAL.md` with `profile: software|data`, the mission `$goal`, definition of
done, and constraints supplied by the user. Data goals must also state ground
truth, reconciliation tolerance, key uniqueness, and rerun/idempotence.
- `STATE.md` with `schema: 1`, iteration 0, `max_iterations: 10`, ready status, a mission
fingerprint, rejected approaches, and key decisions.
- `LOG.md` containing `# Trio loop log`.
- Empty `PLAN.md`, `REPORT.md`, and `VERDICT.md`.
The mailbox is runtime evidence and must not be committed. Do not put
credentials or private machine paths in it. After initialization, tell the user
to run `/skill:trio` for one supervised iteration.
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!