What is firepass-trio?

A coding workflow that sends one task through a researcher, an implementer, and a reviewer. If the reviewer finds problems, the implementer can make a limited round of fixes and be reviewed again.

Use it for feature work or bug fixes that need research before implementation and an audit afterward. It is not intended for research-only, implementation-only, or review-only tasks.

It combines planning, coding, and checking in one workflow, reducing the need to coordinate separate agent calls yourself. It is intended for changes that benefit from both a plan and a review.

firepass-trio — Skill for Claude Code and Codex by jameshgrn

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.

Any agent

npx skills add jameshgrn/firepass-mcp --skill firepass-trio

Clone the repo

git clone --depth 1 https://github.com/jameshgrn/firepass-mcp

Made for: Claude Code, Codex.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/jameshgrn/firepass-mcp/firepass-trio/github.svg)](https://agentmods.dev/skills/jameshgrn/firepass-mcp/firepass-trio)

Your own site

<a href="https://agentmods.dev/skills/jameshgrn/firepass-mcp/firepass-trio"><img src="https://agentmods.dev/badge/skills/jameshgrn/firepass-mcp/firepass-trio/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/jameshgrn/firepass-mcp/firepass-trio"><img src="https://agentmods.dev/badge/skills/jameshgrn/firepass-mcp/firepass-trio.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 79 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 623 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.

Model	Per session	Once invoked
Fable 5.1	$0.00079	$0.00623
Opus 5.5	$0.00032	$0.00249
Sonnet 5.5	$0.00016	$0.00125
Haiku 4.5	$0.00008	$0.00062

Measured 2mo ago against content hash 5eeefbb597a5, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

firepass-trio 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 2mo 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.

skills/firepass-trio/SKILL.md · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

FirePass Trio

Autonomous end-to-end agent powered by Kimi K2.6 Turbo. Chains researcher → worker → reviewer in a single MCP call. If the reviewer finds issues and marks them NEEDS-FIXES, the trio loops back: the worker re-runs with reviewer feedback appended to context, then the reviewer re-runs. The loop is bounded by max_review_rounds.

When to use

End-to-end feature implementation with built-in audit
Bug fixes that should be planned before coded
Any change you'd otherwise call all three single-role tools for

When NOT to use

Pure research (use firepass-researcher)
Pure implementation with no plan needed (use firepass-worker)
Pure review of an existing diff (use firepass-reviewer)

Preference

Use INSTEAD OF chaining the three single-role tools manually.

How to use

Call mcp__firepass__firepass_trio with:

prompt: Clear description of the coding task
cwd: Important — set to the project root so the agent can find files
context: Optional pre-loaded file contents, error messages, or specs
max_iterations: Tool-call rounds per sub-agent (default 60, capped at 200)
max_review_rounds: Fix loop rounds between worker and reviewer (default 2, capped at 5)

Response shape

<firepass_trio status="..." rounds="N">
  <research>...</research>
  <rounds>
    <round n="1">
      <implementation>...</implementation>
      <review>...</review>
    </round>
    ...
  </rounds>
</firepass_trio>

Status values

approved: Review passed with no blocking issues, or all issues were fixed within the review rounds
needs_fixes: Reviewer marked NEEDS-FIXES but max review rounds were exhausted
research_failed: The researcher step did not complete successfully
implementation_failed: The worker step did not complete successfully
review_failed: The reviewer step did not complete successfully

Security

Researcher and reviewer are read-only; worker has full access including bash. All I/O is sandboxed to the provided cwd. See the per-role tool lists in skills/firepass-researcher/SKILL.md, skills/firepass-worker/SKILL.md, and skills/firepass-reviewer/SKILL.md.

Read the full file on GitHub · 68 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.

2mo ago First seen · 68 lines · 79 tokens per session scan A 5eeefbb597a5

Subscribe to this mod's changes

firepass-trio is a skill published in the GitHub repository jameshgrn/firepass-mcp (2 stars, last pushed 5mo ago), licensed MIT. It adds 79 tokens to every session and 623 once invoked, about $0.0003 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 2026-08-31.

Building agents

firepass-trio next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
firepass-triojameshgrn/firepass-mcp	Skill	2	5mo ago	MIT	A	79	Claude Code, Codex
imagegenopenai/codex	Skill	134k	today	Apache-2.0	A	113	Codex
local-ai-agentsmicrosoft/ai-agents-for-beginners	Skill	79k	2d ago	MIT	A	200	Claude Code, Codex
playwright-climicrosoft/playwright	Skill	100k	yesterday	Apache-2.0	A	19	Claude Code
browser-automationopenclaw/openclaw	Skill	394k	today	MIT	A	36	Claude Code
systematic-debuggingobra/superpowers	Skill	315k	5d ago	MIT	A	21	Claude Code
next-cache-components-adoptionvercel/next.js	Skill	143k	today	MIT	A	95	Claude Code

firepass-trio

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

FirePass Trio

When to use

When NOT to use

Preference

How to use

Response shape

Status values

Security

What this file has done since we first saw it

firepass-trio compared with skills from other repositories