A review panel that creates three to five independent reviewers to re-examine a design, plan, or decision from different viewpoints, then combines their conclusions.

Use it for adversarial reviews of plans, designs, or important choices during an agent session.

It helps expose overlooked risks, disagreements, and weak assumptions before you act on a decision.

braintrust — Plugin for Claude Code by cubha/claude-workflow-plugins

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.

Clone the repo

git clone --depth 1 https://github.com/cubha/claude-workflow-plugins

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.

README.md

[![agentmods](https://agentmods.dev/badge/plugins/cubha/claude-workflow-plugins/braintrust/github.svg)](https://agentmods.dev/plugins/cubha/claude-workflow-plugins/braintrust)

Your own site

<a href="https://agentmods.dev/plugins/cubha/claude-workflow-plugins/braintrust"><img src="https://agentmods.dev/badge/plugins/cubha/claude-workflow-plugins/braintrust/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/plugins/cubha/claude-workflow-plugins/braintrust"><img src="https://agentmods.dev/badge/plugins/cubha/claude-workflow-plugins/braintrust.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session not measured What this adds to a session before it is invoked.

When invoked not measured The manifest loads nothing itself; its 1 skill cost 215 tokens a session between them.

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.

Security

Grade A, and why

braintrust 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 1mo 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.

plugins/braintrust/.claude-plugin/plugin.json · 9 lines

What it actually says

{
  "name": "braintrust",
  "version": "1.0.0",
  "description": "In-session adversarial review panel: 3-5 independent fork agents re-examine the session's design, plan, or decision through different lenses, then reconcile conflicts into a verdict. A context-inheriting red team.",
  "author": {
    "name": "cubha"
  }
}

Contents

What it installs

The manifest is a name and a version. 1 skill travel with it, and installing the plugin installs all of them — 215 tokens a session between them. Each is measured on its own page, and each can be installed alone.

Skill braintrust A 215 tokens

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.

1mo ago First seen · 9 lines scan A d222c7feadbc

Subscribe to this mod's changes

braintrust is a plugin published in the GitHub repository cubha/claude-workflow-plugins (2 stars, last pushed 2mo ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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 Claude Code Planning ai-agents claude-code-plugin claude-code-skills claude-plugin-marketplace code-analysis

code-craftsmanship

Code quality and software design skills including Clean Code (Robert C. Martin), Refactoring patterns (Martin Fowler), A Philosophy of Software Design (John Ousterhout), The Pragmatic Programmer (Hunt & Thomas), Domain-Driven Design (Eric Evans), Working Effectively with Legacy Code (Michael Feathers), and Google's…

wondelai/skills · not measured

shinkoku

A Claude Code plugin for Japanese tax filing. It supports employees with side businesses, including business-income, blue-return, income-tax, and consumption-tax filings.

kazukinagata/shinkoku · not measured

ui-clone-skills

Motion forensics for the animated web. Four jobs against a live URL: decode (analyse motion + build, no code), clone (URL → React + Tailwind), verify (AE/SSIM + motion-parity score vs. an existing impl), extract (raw JSON dump). Extracts real CSS and animation params (GSAP / Framer Motion / Lenis / anime.js) directly…

voidmatcha/ui-clone-skills · not measured

pocketto

Pocket-driven development skills: structured subagent delegation, bug hunting, iterative planning, code review workflows, and an opt-in education mentor mode for Claude Code.

rfxlamia/pocketto · not measured

dekko

Static code map for LLM agents. /map parses the repository with tree-sitter into MAP.md and map.json (every file, function, signature, type, and call edge) without spending model tokens; an 18-tool MCP server (searchcode, outline, querysymbol, getcallers, workset, impactedtests, ...) answers structural questions fr.

aahlijia/dekko · not measured

sprint

Orchestrate staged milestones with native subagents on Claude Code or Oh My Pi, plus codex or mimo executors on Claude Code.

onsails/skills · not measured