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/daidaij/websearch-mcpserver/agents-mdgit clone --depth 1 https://github.com/daidaiJ/websearch-mcpserverWrote 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/daidaij/websearch-mcpserver/agents-md)<a href="https://agentmods.dev/instructions/daidaij/websearch-mcpserver/agents-md"><img src="https://agentmods.dev/badge/instructions/daidaij/websearch-mcpserver/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 | $0.01872 | $0.01872 |
| Opus 5 | $0.00936 | $0.00936 |
| Sonnet 5 | $0.00374 | $0.00374 |
| Haiku 4.5 | $0.00187 | $0.00187 |
Grade A, and why
websearch-mcpserver 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — 智能体协作指南
本文件帮助 AI 智能体快速理解项目结构和开发约定，确保长期维护与持续开发的一致性。
项目一句话定位
websearch-mcpserver 是一个用 Go 编写的轻量级 MCP 搜索服务，零 API Key 即可运行，支持 Claude Code / Qwen Code / Cursor 等 MCP 客户端。提供四大核心能力：
| 能力 | 说明 |
|---|---|
| 搜索引擎搜索 | 内置百度/Bing/DuckDuckGo/Google 多引擎并发编排，支持 Tavily/Exa 等 API 引擎混合 |
| API 搜索 | API Key 池轮转模式，支持百度千帆/Tavily/Exa 等多种 API 供应商，失败自动切换 |
| 学术搜索 | arXiv/Crossref/OpenAlex/PubMed/Semantic Scholar/Google Scholar/Europe PMC/DBLP/DOAJ 九大学术引擎并发，DOI 跨引擎去重、逐引擎错误透传 |
| 网页抓取 | 增强型网页内容提取（TLS 指纹伪装 + SSRF 防护 + Jina Reader 备选） |
| PDF 解析 | MinerU AI 增强 PDF 解析（表格/公式/多栏/图片智能识别），无 Token 自动降级 |
技术栈速览
| 项 | 值 |
|---|---|
| 语言 | Go 1.26+，纯 Go 无 CGO |
| 数据库 | SQLite（modernc.org/sqlite，纯 Go 实现） |
| 配置 | Viper（YAML + 环境变量覆盖） |
| 日志 | Zerolog 结构化日志 |
| MCP 协议 | modelcontextprotocol/go-sdk |
| HTTP 客户端 | resty.dev/v3 + go-webfetch（TLS 指纹伪装） |
| 部署 | 单二进制 / Docker 多阶段构建 |
目录结构与模块职责
cmd/ # 入口：main.go + 平台初始化（Windows 代理检测等）
mcp/ # MCP 协议层：工具注册、请求处理
server/ # HTTP 服务：生命周期、路由
searxng/ # SearXNG 兼容 HTTP 端点
pkg/
├── search/ # ★ 核心编排层
│ ├── inf.go # 接口定义（SearchInf, SearchResult）
│ ├── hybrid.go # 多引擎并发编排、去重、合并、排序
│ ├── factory.go # 引擎工厂：根据配置选择并实例化引擎
│ ├── engine_adapter.go # 通用引擎适配器（Tavily/Exa/Google/DDG）
│ ├── baidu_fallback.go # 百度适配器（含智能回退）
│ ├── bing_adapter.go # Bing 适配器
│ ├── apipool.go # API Key 池轮转
│ └── *_test.go # 单元测试
├── antirobot/ # 反检测公共层：Searcher 接口、限流器、TLS 指纹
├── baidu/ # 百度底层引擎实现
├── bing/ # Bing 底层引擎实现
├── ddg/ # DuckDuckGo 底层引擎实现
├── google/ # Google 底层引擎实现
├── academic/ # 学术搜索：arXiv/Crossref/OpenAlex/PubMed/S2/GS/EuropePMC/DBLP/DOAJ
├── config/ # 配置加载与结构体定义
├── cache/ # SQLite 缓存（6h 过期，后台清理）
├── webfetch/ # 增强型网页抓取（SSRF 防护、DNS rebinding 检测）
├── jina/ # Jina Reader 备选抓取
├── llm/ # LLM 摘要生成
├── mineru/ # MinerU PDF 解析
├── proxy/ # 系统代理自动检测（Windows 注册表 / 环境变量）
├── daemon/ # 引用计数进程管理
├── log/ # 日志配置
└── xml/ # XML 格式化
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.
- 4d ago First seen · 146 lines · 1,872 tokens per session scan A 8d4c2aac1240
websearch-mcpserver AGENTS.md is an instructions file published in the GitHub repository daidaiJ/websearch-mcpserver (20 stars, last pushed 4d ago), licensed MIT. It adds 1,872 tokens to every session, about $0.0094 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.
Other instructions, from other repositories
openalex-mcp-server AGENTS.md
AGENTS.md instructions for cyanheads/openalex-mcp-server, covering developer protocol, domain, mcp surface, config and what's next?.
openalex-mcp-server CLAUDE.md
Claude Code instructions for cyanheads/openalex-mcp-server, covering developer protocol, domain, mcp surface, config and what's next?.
academia_mcp CLAUDE.md
Instructions for IlyaGusev/academia_mcp, covering claude.md, project overview, development commands, setup and create virtual environment and install dependencies.
blockrun-mcp AGENTS.md
AGENTS.md instructions for BlockRunAI/blockrun-mcp, covering blockrun mcp, commands, project structure, key dependencies and install in codex.
openrouter-mcp-multimodal AGENTS.md
AGENTS.md instructions for stabgan/openrouter-mcp-multimodal, covering agent instructions, before you ship, releasing (read this before publishing), short version and version files (must all match package.json).
intervals-icu-mcp CLAUDE.md
Instructions for hhopke/intervals-icu-mcp, covering claude.md, project overview, development commands, architecture (quick reference) and tool categories.