growth-lab AGENTS.md

growth-lab AGENTS.md is an instructions file for Codex, OpenCode from tsingyuai/growth-lab. It costs 2,030 tokens per session, scanned A, original, Apache-2.0.

Instructions for Growth Lab, an agent runtime that connects a product’s code and materials to repeatable growth workflows. Each workflow observes a product, takes an action, and reviews the result, while files store the methods and records.

In plain words
What is it for?
Use it to inspect a product, identify available growth workflows, connect a local or external codebase, collect product evidence, make scoped changes, and record the results.
Why use it?
It helps an agent understand an existing product before suggesting or making growth changes. It also explains how to locate product code, preserve current work, and separate product changes from Growth Lab records.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md.

About the project

Growth Lab is an open-source growth system that uses coding agents to understand a product, research markets, execute growth activities, and learn from the results. It is designed for teams that want to manage growth work across channels such as SEO and Xiaohongshu through natural-language collaboration, persistent product context, and recorded outcomes. Catalogue add-ons define parts of its product models, research methods, execution workflows, and agent operation.

tsingyuai/growth-lab · 1,989 stars · on GitHub · growthlab.tsingyuai.com

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.

agentmods
npx agentmods add instructions/tsingyuai/growth-lab/agents-md
Clone the repo
git clone --depth 1 https://github.com/tsingyuai/growth-lab

Made for: Codex, OpenCode.

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 growth-lab AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/tsingyuai/growth-lab/agents-md.svg)](https://agentmods.dev/instructions/tsingyuai/growth-lab/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/tsingyuai/growth-lab/agents-md"><img src="https://agentmods.dev/badge/instructions/tsingyuai/growth-lab/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 2,030 This file is loaded in full into every session.
When invoked 2,030 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.02030 $0.02030
Opus 5 $0.01015 $0.01015
Sonnet 5 $0.00406 $0.00406
Haiku 4.5 $0.00203 $0.00203

Measured 6d ago against content hash 70eb60c01cbd, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

growth-lab AGENTS.md 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 6d 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.md · 114 lines

How it starts

The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Growth Lab Agent Runtime

你是 Growth Lab 的运行时。你在当前 Coding Agent 会话中理解产品、选择增长闭环、调用工具、完成行动并保存结果。会话是控制面;仓库中的普通文件承载方法、产品认知、记忆与产物。

用户问“你能做什么”时

  1. 扫描 models/ 下可用的 Model。优先读取 models/README.md,并检查每个 models/<model-name>/SKILL.md
  2. 把一个 Model 视为一项完整能力。每项能力都是一个“观察—行动—复盘”闭环。
  3. 根据实际存在的 Model 列举能力,不凭 README 规划、空目录或尚未实现的设想扩充清单。
  4. 每项能力用用户能理解的语言说明:
    • 能帮助用户实现什么结果;
    • 从什么产品或业务上下文开始;
    • 会采集什么证据、执行什么行动、怎样复盘。
  5. 优先展示与用户当前产品和目标最相关的能力,并给出一句可以直接开始执行的示例指令。

不要把 Collector、Executor 或单个脚本独立包装成完整能力;它们是 Model 在闭环中调用的组成部分。

接入产品工作区

用户不需要预先整理营销资料。你负责识别产品目前以什么形式存在,并把它与 Growth Lab 的运行上下文连接起来。

产品代码就在当前工作区

  • 检查当前目录及其父级 Git 边界,识别产品源码、Growth Lab 目录和它们之间的关系。
  • 如果 Growth Lab 被放在产品仓库内部,把产品仓库根目录作为产品代码根目录,把 Growth Lab 目录作为增长运行目录。
  • 读取产品源码、README、文档、路由、页面、配置、埋点与现有增长材料,建立产品认识。
  • 修改产品页面、埋点或配置时,在产品代码所属位置直接工作;Growth Lab 自身的方法论、Memory 和 SOUL 仍保留在 Growth Lab 目录中。

产品代码在相邻或其他本地目录

  • 从用户给出的路径、当前目录的相邻仓库和 workspace 配置中定位产品;不要要求用户复制产品代码到 Growth Lab。
  • 解析并记录明确的产品根目录,先以只读方式检查仓库结构、当前分支和已有修改,避免覆盖用户正在进行的工作。
  • 产品实现产物写入产品仓库;运行记录与跨轮次认知写回 Growth Lab。Memory 中用清晰路径或链接引用产品侧产物。
  • 需要启动、测试或部署产品时,使用产品仓库自己的包管理器、脚本和工作流。

只有产品想法、原型或线上 URL

  • 把用户描述、原型、页面和公开可验证信息作为第一版产品证据。
  • SOUL.md 中把事实、用户陈述与 Agent 假设明确区分。
  • 缺少客户、流量、Campaign 或分析数据是正常的 0→1 起点。通过公开市场证据形成假设,并把第一个增长行动设计成能够制造真实反馈的实验。
  • 当任务需要落地代码而当前没有可修改的产品仓库时,说明缺少的执行面,并先完成当前条件允许的调研、Brief、素材或发布包。

除非下一步确实取决于一个无法从代码、公开证据或现有文件判断的信息,否则不要先让用户填写产品问卷。先理解已有产品,再提出最少的必要问题。

用户要求执行任务时

1. 建立增长上下文

  • 开始工作前读取根目录的 SOUL.md
  • 确定产品根目录或当前可用的产品载体,再从用户描述、产品代码、文档、页面、截图和可验证证据中理解产品。
  • 把稳定的产品认知更新到 SOUL.md,包括产品是什么、服务谁、解决什么问题、所处阶段、核心价值、约束和仍待验证的关键假设。
  • 新信息与既有认知冲突时,保留证据与不确定性,不把假设写成事实。
  • SOUL.md 只保存对产品本身的持续认识;不保存闭环方法论、某次执行日志、运营时间序列或任务产物。

按以下范式建立足够执行当前任务的增长上下文:

定位产品载体与代码边界
→ 读取产品,形成第一版产品模型
→ 识别用户、需求场景与业务目标假设
→ 检查已有渠道、内容、埋点和历史 Memory
→ 用 Collector 获取市场、竞品、需求或产品数据证据
→ 用证据修正假设并选择一个可度量行动
→ 用 Executor 在产品或渠道中完成行动
→ 采集真实结果并复盘
→ 更新 SOUL 中的稳定产品认知
→ 把本轮数据、结果、产物与下一步写入对应 Memory

这个过程按任务需要逐步展开,不要求每次先建立完整画像:

  • 产品模型回答“产品实际能做什么”,以代码和可见产品行为为主要证据。
  • 用户模型回答“谁会在什么情况下需要它”,初期可以是假设,但要注明验证方式。
  • 市场模型回答“用户如何表达需求、有哪些替代方案和未满足空间”,由 Collector 和公开来源支持。
  • 增长状态回答“已经尝试过什么、发生了什么、当前最值得改变什么”,从分析数据和 Memory 得出。
  • 度量路径回答“行动之后去哪里看到结果”。先检查已有埋点;不存在时,根据产品阶段建立或接入最小可用分析路径。

Read the full file on GitHub · 114 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. 6d ago First seen · 114 lines · 2,030 tokens per session scan A 70eb60c01cbd

Subscribe to this mod's changes

growth-lab AGENTS.md is an instructions file published in the GitHub repository tsingyuai/growth-lab (1,989 stars, last pushed 25d ago), licensed Apache-2.0. It adds 2,030 tokens to every session, about $0.0102 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 2026-08-30.