trio-agent-loop: Skill for Claude Code

.claude/skills/trio-ship/SKILL.md

trio-ship is a skill for Claude Code from albiol2004/trio-agent-loop. It costs 0 tokens per session (453 once invoked), scanned A, original, MIT.

A procedure for completing an orphaned SHIP verdict, where SHIP means the evaluator approved the work but the expected Git commits are missing.

In plain words
What is it for?
Use it when loop/VERDICT.md contains a machine-readable SHIP verdict whose commit lines were not created.
Why use it?
It restores the product and mailbox records needed to close a verdict that was approved without recording its commits.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions subagents.

This is albiol2004/trio-agent-loop's own configuration. It tells Claude Code how to work on trio-agent-loop itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything trio-agent-loop configures →

Reuse

Borrowing it

Nothing to install: this file belongs to albiol2004/trio-agent-loop. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/albiol2004/trio-agent-loop/main/.claude/skills/trio-ship/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/albiol2004/trio-agent-loop

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for trio-ship

README.md
[![agentmods](https://agentmods.dev/badge/skills/albiol2004/trio-agent-loop/trio-ship.svg)](https://agentmods.dev/skills/albiol2004/trio-agent-loop/trio-ship)
Your own site
<a href="https://agentmods.dev/skills/albiol2004/trio-agent-loop/trio-ship"><img src="https://agentmods.dev/badge/skills/albiol2004/trio-agent-loop/trio-ship.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 453 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00000 $0.00453
Opus 5 $0.00000 $0.00227
Sonnet 5 $0.00000 $0.00091
Haiku 4.5 $0.00000 $0.00045

Measured 8d ago against content hash 04ae78bc231a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

trio-ship 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 8d 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.

.claude/skills/trio-ship/SKILL.md · 34 lines

What it actually says

A SHIP verdict normally ends with the Evaluator's retirement commit: one product commit slice(<id>): <summary> plus one mailbox commit loop: iteration N — SHIP, recorded as commit: lines in loop/VERDICT.md. If a verdict says SHIP but those commits never landed (a legacy or orphaned run), recover it yourself — the same two-commit pattern, using the verdict's own suggested commit message. No subagent and no model config: you run the git commands directly.

Mailbox directory: default loop/. If invoked with dir=<path>, that directory is the mailbox — every loop/ reference below means it.

  1. Read loop/VERDICT.md. Its first non-empty line must be VERDICT: SHIP; otherwise STOP and tell the user the verdict is not machine-readable.
  2. Take the suggested commit message from the verdict's ## Guidance for next iteration section.
  3. Check git status. Clean tree → there is no product commit to make; record commit: <HEAD sha> and jump to step 6.
  4. Commit the product changes attributable to the loop in one commit: git add <those paths> then git commit -m "slice(<id>): <summary>" (the suggested message, shaped to the slice convention; other slice ids in the body). Leave unrelated changes uncommitted and flag them to the user — never sweep foreign paths in.
  5. Append commit: <product sha> to loop/VERDICT.md.
  6. Commit the mailbox: git add loop/ then git commit -m "loop: iteration N — SHIP" (iteration N from STATE.md).
  7. Confirm and report both shas to the user (git rev-parse the product commit and the mailbox commit).
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. 8d ago First seen · 34 lines · 0 tokens per session scan A 04ae78bc231a

Subscribe to this mod's changes

trio-ship is a skill published in the GitHub repository albiol2004/trio-agent-loop (2 stars, last pushed 8d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 453 tokens. 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 2026-08-31.