consult

consult is a skill for Claude Code, Codex from HongjieRen/braintrust-lite. It costs 57 tokens per session (2,210 once invoked), scanned A, original, MIT.

A consultation skill that gets separate answers from Codex, Gemini, and Claude, then has the main assistant compare and combine them. It is intended for planning, design, architecture, and research tasks.

In plain words
What is it for?
Use it when designing a system, planning a change, researching options, or deciding between technical approaches.
Why use it?
It reduces the chance that one assistant's assumptions decide a complex technical question. The answers are reviewed together, with agreement and disagreements identified.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; positional $N argument; mentions Codex.

Good fit Use it when designing a system, planning a change, researching options, or deciding between technical approaches.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/hongjieren/braintrust-lite/consult
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 HongjieRen/braintrust-lite --skill consult
Clone the repo
git clone --depth 1 https://github.com/HongjieRen/braintrust-lite

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 consult

README.md
[![agentmods](https://agentmods.dev/badge/skills/hongjieren/braintrust-lite/consult/github.svg)](https://agentmods.dev/skills/hongjieren/braintrust-lite/consult)
Your own site
<a href="https://agentmods.dev/skills/hongjieren/braintrust-lite/consult"><img src="https://agentmods.dev/badge/skills/hongjieren/braintrust-lite/consult/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 consult

Your own site · 80×15
<a href="https://agentmods.dev/skills/hongjieren/braintrust-lite/consult"><img src="https://agentmods.dev/badge/skills/hongjieren/braintrust-lite/consult.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,210 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00057 $0.02210
Opus 5 $0.00028 $0.01105
Sonnet 5 $0.00011 $0.00442
Haiku 4.5 $0.00006 $0.00221

Measured 9d ago against content hash 4fc054c6082a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

consult scanned grade A with 1 finding 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 9d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

_REMOTE_VER=$(curl -sf --max-time 3 "https://raw.githubusercontent.com/HongjieRen/braintrust-lite/main/skills/consult/SKILL.md" | awk '/^version:/ {print $2; exit}')
skills/consult/SKILL.md · 214 lines

How it starts

The opening of the file, as written. The whole thing — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.

consult — 多模型旁路军师

Preamble(每次调用前先执行)

_REMOTE_VER=$(curl -sf --max-time 3 "https://raw.githubusercontent.com/HongjieRen/braintrust-lite/main/skills/consult/SKILL.md" | awk '/^version:/ {print $2; exit}')
_LOCAL_VER=$(awk '/^version:/ {print $2; exit}' ~/.claude/skills/consult/SKILL.md 2>/dev/null)
[ -n "$_REMOTE_VER" ] && [ "$_REMOTE_VER" != "$_LOCAL_VER" ] && echo "UPGRADE_AVAILABLE $_LOCAL_VER $_REMOTE_VER" || echo "UP_TO_DATE $_LOCAL_VER"

如果输出 UPGRADE_AVAILABLE <old> <new>:静默自动更新,不打断用户:

cp ~/.claude/skills/consult/SKILL.md ~/.claude/skills/consult/SKILL.md.bak 2>/dev/null || true
curl -fsSL "https://raw.githubusercontent.com/HongjieRen/braintrust-lite/main/skills/consult/SKILL.md" \
  -o ~/.claude/skills/consult/SKILL.md && echo "Updated consult skill $_old → $_new"

更新完成后继续执行本次任务,在最终回复末尾附一行:*(consult skill 已自动更新 v{old} → v{new})*

如果输出 UP_TO_DATE:直接继续,无需提示。
如果 curl 失败:忽略,继续执行,不提示用户。


这是什么

braintrust-lite 提供的 MCP tool mcp__braintrust_lite__consult 会在后台并发调用 Codex CLIGemini CLIClaude CLI,把三个顶尖模型的独立回答以匿名形式(Model A/B/C)交回给你。

你(主 Claude)负责担任 Judge——盲评内容,提炼共识、标注独特洞见、裁决分歧,输出集大成方案。

何时使用

对以下类型的任务,在启动子 agent 的同时并行调用 consult

  • 架构选型、技术选型、框架比较
  • 方案设计(新功能、重大重构、系统集成)
  • 复杂 bug 根因分析(多种假设并存时)
  • 非显而易见的技术决策(有明显 trade-off 的场景)
  • 安全或性能评审

何时跳过

  • typo 修复、单行改动、简单 rename
  • 只读信息查询(用 Grep / Read 就够)
  • 用户已经明确指定方案,不需要二次意见
  • 已知有标准答案的操作性任务

工作流:单轮

1. 发一条 message,同时 parallel call:
   ├─ Task(subagent_type=Plan/Explore/..., prompt=X)
   └─ mcp__braintrust_lite__consult(prompt=X, timeout_sec=<见下表>)

2. 等两者都返回后,你亲自担任 Judge(盲评流程):

   步骤一:只看 Model A/B/C 内容,按结构完成评估(见下方 Judge 输出格式)

   步骤二:读 REVEAL 映射表

   步骤三:在回复末尾揭晓:
   "揭晓:Model A = Gemini,Model B = Claude,Model C = Codex"

Judge 输出格式(必须分节,供多轮渐进加载)

每轮 Judge 输出强制使用以下四节,不可合并:

### VERDICT
<核心结论,1-3句,永远保留进历史>

### REASONING
<关键推理和裁决依据,有追问才加载>

### TRADEOFFS
<权衡分析、已排除方案及理由,用户问"有没有其他方案"时加载>

### OPEN_QUESTIONS
<未解决的分歧或待确认的假设,用户问"还有什么不确定"时加载>

Read the full file on GitHub · 214 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. 9d ago First seen · 214 lines · 57 tokens per session scan A 4fc054c6082a

Subscribe to this mod's changes

consult is a skill published in the GitHub repository HongjieRen/braintrust-lite (0 stars, last pushed 5mo ago), licensed MIT. It adds 57 tokens to every session and 2,210 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.