agentmods.dev

bro-mcp-loop

bro-mcp-loop is a skill for Claude Code, Codex from wedabro/bro-skills. It costs 81 tokens per session (1,107 once invoked), scanned A, original, MIT.

An automated workflow for coordinating several coding agents through planning, implementation, testing, and review. It sends work between a planner, an executor, and reviewers, then repeats fixes when review finds problems.

In plain words
What is it for?
Use it to plan a feature, assign coding work, run automated tests, perform code and interface reviews, and address reported bugs until the work meets its criteria.
Why use it?
It removes the need to coordinate each agent and review step manually, while keeping testing and independent review in the workflow.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents); mentions Codex.

not rated 3repo 8d ago A scan 81 tokens original MIT

Good fit Use it to plan a feature, assign coding work, run automated tests, perform code and interface reviews, and address reported bugs until the work meets its criteria.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/wedabro/bro-skills/bro-mcp-loop
View source ↗ wedabro/bro-skills
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 wedabro/bro-skills --skill bro-mcp-loop
Clone the repo
git clone --depth 1 https://github.com/wedabro/bro-skills

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.

agentmods badge for bro-mcp-loop

README.md
[![agentmods](https://agentmods.dev/badge/skills/wedabro/bro-skills/bro-mcp-loop/github.svg)](https://agentmods.dev/skills/wedabro/bro-skills/bro-mcp-loop)
Your own site
<a href="https://agentmods.dev/skills/wedabro/bro-skills/bro-mcp-loop"><img src="https://agentmods.dev/badge/skills/wedabro/bro-skills/bro-mcp-loop/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.

agentmods 80×15 button for bro-mcp-loop

Your own site · 80×15
<a href="https://agentmods.dev/skills/wedabro/bro-skills/bro-mcp-loop"><img src="https://agentmods.dev/badge/skills/wedabro/bro-skills/bro-mcp-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 81 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,107 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 $0.00081 $0.01107
Opus 5.5 $0.00032 $0.00443
Sonnet 5.5 $0.00016 $0.00221
Haiku 4.5 $0.00008 $0.00111

Measured 27d ago against content hash 03ba5d8bf84c, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

bro-mcp-loop 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 27d 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.

.agents/skills/bro-mcp-loop/SKILL.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

🔁 bro-mcp: Autonomous Multi-Agent Loop Skill

Skill này đóng gói toàn bộ quy trình phối hợp khép kín giữa các AI Agent mà người dùng không cần phải nhớ hay gọi thủ công từng MCP tool.

🎯 Khi Nào Kích Hoạt Skill Này?

  • Khi người dùng yêu cầu: "Chạy loop làm tính năng X", "Dùng bro-mcp", "Giao cho Antigravity code rồi Codex review", hoặc gõ /bro-mcp-loop.
  • Khi cần phát triển tính năng mới với yêu cầu nghiêm ngặt: Có kiểm thử tự động, có review độc lập và tự động sửa bug khi có phản hồi.

🚦 Quy Trình Vận Hành 4 Bước Tự Động (Auto-Orchestration)

Bước 1: Kiểm Tra Kết Nối (Pre-flight Check)

  • Xác nhận server bro-mcp đang hoạt động tại http://localhost:3088/health.
  • Nếu server chưa chạy, nhắc người dùng bật container: docker compose up -d tại f:\wedabro\bro-mcp.

Bước 2: Vai Trò Lead Planner (Codex / Cursor) — Lập Kế Hoạch & Giao Task

  1. Phân tích yêu cầu của người dùng, phân rã WBS và xác định kiến trúc.
  2. Thiết lập 2 - 4 tiêu chí Acceptance Criteria (AC) rõ ràng dạng Given-When-Then.
  3. Tự động gọi tool MCP: 
    bro_dispatch_task({
  "title": "<Tên tính năng ngắn gọn>",
  "description": "<Mô tả kiến trúc và các ràng buộc kỹ thuật>",
  "acceptanceCriteria": [
    "AC1: <Tiêu chí bắt buộc 1>",
    "AC2: <Tiêu chí bắt buộc 2>"
  ],
  "assignedTo": "antigravity",
  "maxIterations": 3
})
  4. Báo cáo cho người dùng mã taskId đã được giao thành công ở trạng thái DISPATCHED.

Bước 3: Vai Trò Executor (Antigravity) — Nhận Việc & Thi Công

  1. Tự động gọi tool MCP: 
    bro_fetch_task({ "taskId": "<taskId>" })
  2. Đọc toàn bộ specs, tiêu chí AC và hướng dẫn sửa lỗi (nếu đang ở vòng lặp revision).
  3. Trực tiếp thực thi:
    • Sửa/tạo file mã nguồn, không để lại placeholder hoặc TODO.
    • Chạy kiểm thử tự động trong Docker hoặc terminal.
    • Chụp ảnh giao diện UI / Generative UI render (nếu có frontend).
  4. Tự động gọi tool MCP nộp sản phẩm: 
    bro_submit_for_review({
  "taskId": "<taskId>",
  "gitDiff": "<toàn bộ mã thay đổi>",
  "testResults": "<log kết quả chạy test tự động>",
  "uiArtifacts": ["<đường dẫn ảnh giao diện nếu có>"],
  "notes": "<tóm tắt các tiêu chí AC đã hoàn thành>"
})
  5. Trạng thái task chuyển sang UNDER_REVIEW.

Read the full file on GitHub · 88 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.

  1. 27d ago First seen · 88 lines · 81 tokens per session scan A 03ba5d8bf84c

Subscribe to this mod's changes

bro-mcp-loop is a skill published in the GitHub repository wedabro/bro-skills (3 stars, last pushed 8d ago), licensed MIT. It adds 81 tokens to every session and 1,107 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 2026-09-24.

Building agentsTestingagentai-agentsclaudeclaude-skillscodex-skillskills
Related

Other skills, from other repositories

random-user-post-batch

Returns multiple randomly generated fake user profiles in a single request. Each call consumes count units of quota. Maximum 50 users per request.

bobadilla-tech/requiems-api-skills · 33 tokens

random-user-get-random-user

Returns a randomly generated fake user profile.

bobadilla-tech/requiems-api-skills · 15 tokens

offensive-fuzzing

Practical offensive fuzzing methodology covering target identification, fuzzer selection (AFL++, libFuzzer, Honggfuzz, Boofuzz, syzkaller), harness writing, corpus curation, mutation strategies, coverage measurement, and crash triage. Use when setting up or running fuzz campaigns against any target: file parsers…

SnailSploit/Claude-Red · 91 tokens

prompt-regression-suite

Design a regression test suite that catches an LLM feature getting worse when the prompt, model, or context changes. Use when asked to stop prompt changes breaking production, set up golden tests or CI gates for an LLM feature, or test a model/prompt upgrade before shipping it. Produces a golden case set, per-case…

mohitagw15856/pm-claude-skills · 102 tokens

ai-agent-reliability

Make an AI agent or automation reliable enough to trust — the tests, checks, and guardrails that catch its failures before they reach anything real. Use when asked how do I test my AI agent, make my automation reliable, my agent works sometimes, or how do I trust an AI workflow in production. Produces a map of where…

mohitagw15856/pm-claude-skills · 135 tokens

ai-eval-plan

Design an evaluation plan for an LLM or AI feature before shipping it. Use when asked how to evaluate a prompt/model/agent, set up an eval harness, define quality metrics for an AI feature, or build a regression gate. Produces an eval plan — task definition, datasets, metrics & rubrics, baselines, automated + human…

mohitagw15856/pm-claude-skills · 86 tokens