think-tank

think-tank is a skill for Claude Code, Codex from yangKJ/think-tank-skill. It costs 45 tokens per session (3,348 once invoked), scanned A, original, MIT.

A coordination skill for routing complex requests across research, discussion, review, and other roles or providers. A provider is the service that performs a delegated task.

In plain words
What is it for?
Collecting information from multiple channels, organising role-based analysis, running deliberations, and combining the results into a conclusion.
Why use it?
It helps decide which capabilities and roles a request needs while keeping permissions and provider boundaries explicit.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; mentions subagents; mentions Claude Code.

Good fit Collecting information from multiple channels, organising role-based analysis, running deliberations, and combining the results into a conclusion.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/yangkj/think-tank-skill/think-tank
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 yangKJ/think-tank-skill --skill think-tank
Clone the repo
git clone --depth 1 https://github.com/yangKJ/think-tank-skill

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 think-tank

README.md
[![agentmods](https://agentmods.dev/badge/skills/yangkj/think-tank-skill/think-tank/github.svg)](https://agentmods.dev/skills/yangkj/think-tank-skill/think-tank)
Your own site
<a href="https://agentmods.dev/skills/yangkj/think-tank-skill/think-tank"><img src="https://agentmods.dev/badge/skills/yangkj/think-tank-skill/think-tank/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 think-tank

Your own site · 80×15
<a href="https://agentmods.dev/skills/yangkj/think-tank-skill/think-tank"><img src="https://agentmods.dev/badge/skills/yangkj/think-tank-skill/think-tank.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 45 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,348 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 $0.00045 $0.03348
Opus 5 $0.00023 $0.01674
Sonnet 5 $0.00009 $0.00670
Haiku 4.5 $0.00005 $0.00335

Measured 11d ago against content hash fb090cac1fc5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

think-tank 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 11d 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.

think-tank/SKILL.md · 393 lines

How it starts

The opening of the file, as written. The whole thing — 393 lines — stays where its author put it; the contents beside it link to each section on GitHub.

think-tank Skill

定位

think-tank 是一个跨平台、可复用的高阶 Skill,用于多角色信息收集、协作分析、讨论审议与结论汇总。

它不是工具合集,不复制外部 skills;它负责任务理解、协议执行、角色组织、能力编排和最终汇总。

Skill 和运行目录边界

think-tank 的 skill 本体只能是当前安装位置中的 SKILL.md、协议、profiles、recipes、routing、platform adapters 和 runtime 文件。

.think-tank/ 是运行配置、provider preflight、runs、memory 和 artifacts 存储目录,不是 skill 目录。不要把 .think-tank/ 放进 .codex/skills/.claude/skills/

provider 边界仍必须遵守:selection 不等于 invocation,真实 provider 调用前必须有明确 dispatch decision、权限确认和结果回收。

3.0 Skill Experience Layer

think-tank 3.0 增加面向 agent 的使用体验层,用来回答三个问题:

  • 这个任务是否应该使用 think-tank?
  • 使用前需要形成什么 invocation contract?
  • 当前任务应该加载哪些最小协议、mode、recipe、profile、platform 或 provider 文档?

先按以下协议做判断:

protocol/skill-trigger-intelligence.md
protocol/skill-invocation-contract.md
protocol/runtime-profile-contract.md
protocol/progressive-disclosure.md

触发词、别名、平台快捷语和 provider 偏好不属于公开 core 的内置规则。它们应由用户自己的 YAML policy 或平台 adapter 定义。公开 core 只能提供 intent 识别原则、trigger category 示例和 policy schema。

当触发来源不明确时,先输出 skill_route_decision,并标注:

policy_source: protocol_default
trigger_status: inferred_intent_only

何时使用

当用户任务满足任一条件时,使用 think-tank:

  • 需要多渠道信息收集
  • 需要多个角色分别判断
  • 需要讨论、审议、观点碰撞或决策
  • 需要审查产物、发现问题或验收
  • 需要策略、路线、产品或架构判断
  • 需要把复杂资料汇总为行动建议

以下只是可被用户 YAML policy 采用的 intent 短语示例,不是 think-tank core 的内置触发规则:

  • 研究:研究一下深度研究全面分析
  • 竞争:竞品分析竞争分析竞品调研
  • 市场:市场调研行业分析用户需求
  • 技术:技术调研方案调研可行性分析
  • 反馈:舆情分析用户反馈评论分析
  • 决策:开会讨论讨论一下
  • 审查:审查review验收找问题
  • 策略:制定策略路线图行动方案

平台无关的 intent 路由见:

protocol/intent-routing.md
protocol/skill-trigger-intelligence.md

跨项目任务配方见:

recipes/

其中 research-to-video 覆盖选题研究、资料调研到视频 brief、分镜、媒体执行记录和质量门禁。

Codex 平台只负责把这些 intent/recipe 映射到当前可用能力,见:

platforms/codex/trigger-routing.md

不应强行使用 think-tank:

  • 简单事实查询
  • 单页摘要
  • 明确的一步命令
  • 不需要多角色或证据判断的任务

执行顺序

1. 解析任务

提取:

  • 用户目标
  • 已知上下文
  • 约束
  • 成功标准
  • 是否需要实时信息
  • 期望输出

Read the full file on GitHub · 393 lines

Files

What ships with it

60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 11d ago First seen · 393 lines · 45 tokens per session scan A fb090cac1fc5

Subscribe to this mod's changes

think-tank is a skill published in the GitHub repository yangKJ/think-tank-skill (2 stars, last pushed 2mo ago), licensed MIT. It adds 45 tokens to every session and 3,348 once invoked, about $0.0002 per session on Opus 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.