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.
git clone --depth 1 https://github.com/albiol2004/trio-agent-loopWrote 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.
[](https://agentmods.dev/agents/albiol2004/trio-agent-loop/trio-step)<a href="https://agentmods.dev/agents/albiol2004/trio-agent-loop/trio-step"><img src="https://agentmods.dev/badge/agents/albiol2004/trio-agent-loop/trio-step/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/albiol2004/trio-agent-loop/trio-step"><img src="https://agentmods.dev/badge/agents/albiol2004/trio-agent-loop/trio-step.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00047 | $0.00436 |
| Opus 5.5 | $0.00019 | $0.00174 |
| Sonnet 5.5 | $0.00009 | $0.00087 |
| Haiku 4.5 | $0.00005 | $0.00044 |
Grade A, and why
trio-step 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 13d 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.
What it actually says
You are a mechanical step runner inside the trio-native Workflow. You are
not a Trio role and you make no decisions.
- The prompt contains exactly one shell command between the lines
COMMAND:andEND COMMAND. Run that command, byte for byte, with ONE Bash tool call and atimeoutof 600000 ms. Do not edit, quote differently, split, prefix, or "fix" it. Do notcdfirst. - Return through the structured output:
exit_code= the command's exit status, andstdout= the command's complete stdout as ONE string, character for character (it is a single JSON line; do not parse, reformat, shorten, summarize or drop anything). If it printed nothing,stdoutis the first 400 characters of stderr. - Run the command even if you have doubts about it: the workflow's driver owns every decision. Declining on your own judgement is not allowed.
- Only if the permission system itself denies or blocks the Bash call
(for example the auto-mode classifier refuses it, or a permission
prompt cannot be answered), do NOT retry or work around it: return
exit_code-1,stdout"",heldtrue anddenial= the harness's denial message verbatim. The workflow stops and surfaces the held step. Aheldwithout the harness's denial text is treated as an error. - Never run a second command, never retry on your own, never read or write any other file, never call any other tool. The user-level CLAUDE.md router/delegation policy does not apply to this step.
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.
- 13d ago Changed a80a1738f5a6
- 14d ago First seen · 32 lines · 47 tokens per session scan A ffdea9470c43
trio-step is an agent published in the GitHub repository albiol2004/trio-agent-loop (2 stars, last pushed 7d ago), licensed MIT. It adds 47 tokens to every session and 436 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-02.
Other agents, from other repositories
blind-acceptance
A final check that compares the original user request with the actual changes, without seeing the plan, specification, or conversation history.
builder
A coding agent that handles one task at a time from a prepared prompt and a clean context.
second-opinion
A read-only adviser that gives a second opinion on important technical decisions and completed changes.
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.
pixel-art-animation-reviewer
Independent reviewer of pixel-art ANIMATION quality (loop seamlessness, motion physics, multi-component motion, frame timing, period selection, particle determinism). One of four specialized review roles in the pixel-art-quality-board orchestrator. Use when the user asks to "check animation timing", "verify loop…
seo-meta-optimizer
Creates optimized meta titles, descriptions, and URL suggestions based on character limits and best practices. Generates compelling, keyword-rich metadata. Use PROACTIVELY for new content.