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.
npx agentmods add instructions/caomeiyouren/tinyfish-mcp/agents-mdgit clone --depth 1 https://github.com/CaoMeiYouRen/tinyfish-mcpWrote 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.
[](https://agentmods.dev/instructions/caomeiyouren/tinyfish-mcp/agents-md)<a href="https://agentmods.dev/instructions/caomeiyouren/tinyfish-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/caomeiyouren/tinyfish-mcp/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00813 | $0.00813 |
| Opus 5 | $0.00407 | $0.00407 |
| Sonnet 5 | $0.00163 | $0.00163 |
| Haiku 4.5 | $0.00081 | $0.00081 |
Grade A, and why
tinyfish-mcp 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 5d 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.
What it actually says
AGENTS.md
项目概述
一个基于 TinyFish 免费搜索 API 的 MCP（Model Context Protocol）服务，为 AI 助手提供实时网络搜索能力。支持通过 Stdio 和 HTTP 两种标准协议进行便捷集成，方便开发者将搜索功能接入到各类 AI 应用中。
角色与目标
- 本文件用于约束 AI 代理与协作者在当前项目中的默认行为。
- 任何更具体的项目规范、目录说明、设计决策或工作流要求，应以项目内的 README、开发规范、设计文档和配置文件为准。
- 当下文内容与项目实际实现冲突时，以代码、配置和项目规范文档中的事实为准。
技术栈
-
主要语言: typescript
-
运行时: nodejs
-
包管理器: npm
仓库信息
- 仓库地址: https://github.com/CaoMeiYouRen/tinyfish-mcp
- 项目文档: https://github.com/CaoMeiYouRen/tinyfish-mcp#readme
- Issue 地址: https://github.com/CaoMeiYouRen/tinyfish-mcp/issues
- 贡献指南: https://github.com/CaoMeiYouRen/tinyfish-mcp/blob/master/CONTRIBUTING.md
项目结构
以下目录为模板默认结构；具体项目可在生成后按实际情况增删。
src/ # 源代码
tests/ # 测试文件
playground/ # 本地调试或演示代码（如有）
docs/ # 文档与规范（如有）
常用命令
- 安装依赖:
npm install - 启动开发环境:
npm run dev - 运行测试:
npm run test - 构建项目:
npm run build - 代码检查:
npm run lint - 启动生产或本地预览:
npm run start - 生成提交:
npm run commit
编码与协作约定
- 优先遵循项目现有代码风格、目录约定和命名规范，不要擅自引入新的体系。
- 使用与项目一致的语言特性和类型策略；如果项目启用了 TypeScript，则保持类型检查可通过。
- 变更应尽量保持最小范围，避免顺手重构无关模块。
- 需要新增或调整依赖时，优先确认项目已有方案是否已覆盖。
- 提交信息遵循 Conventional Commits 规范。
质量门禁
-
测试框架: Vitest
-
目标覆盖率: >= 80%
-
代码变更后，按项目实际可用命令完成 lint、typecheck、test 等必要校验。
-
如果某项校验在当前项目中不存在，应在说明中明确标注，而不是假定其存在。
安全与避免事项
- 不要硬编码 API Key、Token、密码或其他敏感信息。
- 不要直接修改构建产物、发布产物或生成目录中的文件。
- 不要跳过 TypeScript 类型检查或项目规定的静态检查。
- 不要使用
var声明变量，优先使用const和let。 - 对环境变量、密钥文件和部署配置的修改应格外谨慎。
可选补充
- 如果项目存在更细的 AI 规则、目录约定或角色分工，可在此处继续补充，但不要与项目事实相冲突。
- 如果项目已经定义了更强的安全、测试或发布流程，本文件应只保留入口级约束。
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.
- 5d ago First seen · 69 lines · 813 tokens per session scan A 4e04c374ce0c
tinyfish-mcp AGENTS.md is an instructions file published in the GitHub repository CaoMeiYouRen/tinyfish-mcp (2 stars, last pushed 1mo ago), licensed MIT. It adds 813 tokens to every session, about $0.0041 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.
Other instructions, from other repositories
mcp AGENTS.md
AGENTS.md instructions for Teamwork/mcp, covering agents.md, this repository is public, project overview, setup commands and build and run.
better-notion-mcp AGENTS.md
AGENTS.md instructions for n24q02m/better-notion-mcp, covering agents.md - better-notion-mcp, build / lint / test commands, run a single test file, run a single test by name and mise shortcuts.
raindrop-mcp GEMINI.md
Instructions for adeze/raindrop-mcp, covering raindrop mcp server - project context, project overview, core technologies, architecture and key commands.
mcp-dockhand copilot-review.instructions.md
Copilot instructions for strausmann/mcp-dockhand: When reviewing pull requests in this repository.
mcp-dockhand gemini-review.instructions.md
Instructions for strausmann/mcp-dockhand, covering gemini code assist review instructions, upstream verification (critical) and code quality checks.
obsidian-vault-mcp CLAUDE.md
Claude Code instructions for ebullient/obsidian-vault-mcp, covering obsidian vault mcp, code style and qa.