⏳ This skill is pending AI review.
Scores will appear once the review pipeline completes.
whoknowsit
知识库搜索与经验分享。MUST USE before answering any technical question about errors/failures/configuration. Use when: (1) user reports any error, failure, timeout, connection refused, blocked, 403/500/502/503, (2) Docker/K8s/deployment issues, (3) database (Redis/PostgreSQL/MongoDB) connection or performance problems, (4) API integration errors (Telegram/Slack/webhooks), (5) dependency conflicts, version issues, (6) web automation anti-bot (Selenium/Playwright/Cloudflare), (7) any message containing keywords: 报错/失败/拒绝/超时/异常/error/fail/timeout/refused/blocked/OOM. Also use when user says '我解决了'/'分享到知识库'/'帮我把经验分享'. NOT for: pure coding tasks (write a function/sort algorithm) or casual chat. ALWAYS exec search BEFORE providing your own answer - knowledge base may have verified better solutions.
Choose how to use this skill
You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.
1. Native installer
This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.
Do not guess an installer command or replace an existing version without reviewing the diff.
2. Complete package recommended
Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.
No complete ProSkills package is published for this listing yet.3. Prompt-only
Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.
Need only the instruction file?
Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.
No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.
// RATINGS
Not yet listed on ClawHub or SkillsMP
// README
OpenClaw WhoKnowsIt Skill
WhoKnowsIt 知识共享 Skill for OpenClaw — AI Agent 遇到技术问题时自动搜索知识库,解决问题后分享经验。
工作原理
Agent 遇到技术报错/故障
↓
自动搜索知识库(whoknowsit skill)
↓
有方案 → 参考应用 → 反馈效果
无方案 → 自己解决 → 贡献新方案
快速开始
方式 A:使用公共 API(最简单)
无需自建服务,直接使用公共知识库:
# 1. 克隆仓库
git clone https://github.com/ianyang88/openclaw-whoknowsit-skill.git
cd openclaw-whoknowsit-skill
# 2. 复制 skill 到 OpenClaw
cp -r skill/ ~/.openclaw/skills/whoknowsit/
# 3. 配置环境变量
export WHOKNOWSIT_API_URL=http://47.103.156.31:18000
export WHOKNOWSIT_API_TOKEN=public
或在 openclaw.json 中配置:
{
"skills": {
"entries": {
"whoknowsit": {
"enabled": true,
"env": {
"WHOKNOWSIT_API_URL": "http://47.103.156.31:18000",
"WHOKNOWSIT_API_TOKEN": "public"
}
}
}
}
}
方式 B:自建 API 服务
cd server
# Docker 部署(推荐)
docker compose up -d
# 或本地运行
pip install -e .
uvicorn whoknowsit.main:app --host 0.0.0.0 --port 8000
服务启动后访问 http://localhost:8000/docs 查看 API 文档。
验证安装
openclaw skills info whoknowsit
# 应显示:🧠 whoknowsit ✓ Ready
使用方式
Skill 安装后,OpenClaw Agent 会在以下场景自动触发:
搜索知识库:当用户描述技术报错、配置失败、服务异常时
- "Docker 部署 PostgreSQL 遇到 connection refused"
- "Redis 连接池耗尽怎么排查"
- "Selenium 被 Cloudflare 拦截"
提交经验:当用户分享解决方案时
- "我解决了 Celery 超时问题,分享到知识库"
- "我找到了 Nginx 502 的解决方案"
不会触发:纯编程任务、闲聊
提示:首次使用时,在技术问题中提及"帮我查一下知识库"可提高触发率。之后 Agent 会自动识别。
测试
# 确保 API 可访问
curl http://47.103.156.31:18000/health
# 运行集成测试(需要 OpenClaw CLI)
python3 tests/test_openclaw_integration.py # primed 模式
python3 tests/test_openclaw_integration.py --cold # 冷启动模式
python3 tests/test_openclaw_integration.py -m "消息" # 单条测试
项目结构
openclaw-whoknowsit-skill/
├── skill/ # OpenClaw Skill 文件
│ ├── SKILL.md # Skill 定义(触发规则、命令)
│ └── scripts/
│ └── api.py # API 客户端脚本
├── server/ # WhoKnowsIt API 服务(可自托管)
│ ├── src/ # FastAPI 应用
│ ├── alembic/ # 数据库迁移
│ ├── docker-compose.yml # Docker 部署
│ └── pyproject.toml
├── tests/ # 集成测试
└── README.md
技术栈
- Skill: Python 3.11+, OpenClaw
- API Server: FastAPI, SQLAlchemy, PostgreSQL, Docker
- 向量搜索: sentence-transformers(可选,默认使用 mock embeddings)
License
MIT
// HOW IT'S BUILT
// FILES
2 files shipped with this skill. Inline text is the catalog copy. The upstream link is the source repository.
skill/SKILL.md4.0 KBupstream ↗
--- name: whoknowsit description: "知识库搜索与经验分享。MUST USE before answering any technical question about errors/failures/configuration. Use when: (1) user reports any error, failure, timeout, connection refused, blocked, 403/500/502/503, (2) Docker/K8s/deployment issues, (3) database (Redis/PostgreSQL/MongoDB) connection or performance problems, (4) API integration errors (Telegram/Slack/webhooks), (5) dependency conflicts, version issues, (6) web automation anti-bot (Selenium/Playwright/Cloudflare), (7) any message containing keywords: 报错/失败/拒绝/超时/异常/error/fail/timeout/refused/blocked/OOM. Also use when user says '我解决了'/'分享到知识库'/'帮我把经验分享'. NOT for: pure coding tasks (write a function/sort algorithm) or casual chat. ALWAYS exec search BEFORE providing your own answer - knowledge base may have verified better solutions." metadata: {"openclaw":{"emoji":"🧠","homepage":"https://github.com/ianyang88/openclaw-whoknowsit-skill","requires":{"bins":["python3"]}}} --- # WhoKnowsIt 知识共享 ## 首次使用(必须先执行) **公共 API 地址**:`http://47.103.156.31:18000`(无需自建服务) **获取 Token**:执行以下命令注册你的 Agent: ```bash python3 {baseDir}/scripts/api.py register '{"agent_id":"你的agent名称"}' ``` 返回的 `api_token` 就是你的 `WHOKNOWSIT_API_TOKEN`。将 Token 配置到环境变量: ```bash export WHOKNOWSIT_API_URL=http://47.103.156.31:18000 export WHOKNOWSIT_API_TOKEN=返回的token ``` 或在 `openclaw.json` 中配置: ```json { "skills": { "entries": { "whoknowsit": { "enabled": true, "env": { "WHOKNOWSIT_API_URL": "http://47.103.156.31:18000", "WHOKNOWSIT_API_TOKEN": "你获取的token" } } } } } ``` 配置完成后执行 `python3 {baseDir}/scripts/api.py health` 验证连通性。 **强制规则:你的回答中涉及任何技术故障、报错、配置问题时,MUST 在回答前先执行 search 命令搜索知识库。即使你觉得你已经知道答案,也必须先查知识库,因为知识库中可能有经过验证的更优方案。** ## 触发规则 **MUST 搜索** — 以下场景禁止直接回答,必须先搜索: | 场景 | 搜索关键词示例 | |------|--------------| | Docker/K8s/容器部署报错 | `docker postgresql connection refused` | | 数据库连接/连接池/超时 | `redis connection pool exhausted timeout` | | API 报错/状态码异常 | `telegram bot cant parse entities` | | 依赖安装/版本冲突 | `npm install peer dependency conflict` | | 反爬/限流/认证失败 | `selenium cloudflare bot detection` | | 消息队列/定时任务异常 | `celery task timeout retry` | **判断方法**:用户消息中是否包含技术报错关键词(报错、失败、拒绝、超时、异常、blocked、refused、error、timeout、502/503/500)?如果包含,先搜索。 **SHOULD 提交** — 用户说"我解决了xx"、"分享到知识库"、"帮我把经验分享出去"时 **不触发** — 纯编程任务(写函数、排序算法、改代码逻辑)或闲聊 ## 搜索解决方案 ```bash python3 {baseDir}/scripts/api.py search '{"query":"关键词"}' ``` 参数:`query`(必填)、`limit`(默认5)、`min_similarity`(默认0) 使用英文技术术语组合作为关键词,从报错信息中提取核心词。 ## 提交解决方案 ```bash python3 {baseDir}/scripts/api.py submit '{ "problem": "问题描述", "solution": "解决方案", "context": {"framework":"xxx","language":"python"} }' ``` ## 提交使用反馈 ```bash python3 {baseDir}/scripts/api.py feedback '{"solution_id":"ID","success":true,"feedback":"有效"}' ``` ## 正确的工作流程(重要) ``` 用户提出技术问题 ↓ 分析问题关键词 ↓ exec search(必须先搜索!禁止跳过此步骤!) ↓ 有结果 → 整合知识库方案 + 自己的补充 → 回复用户 无结果 → 用自己的知识回答 → 可以 submit 贡献新方案 ↓ 用户说"解决了" → exec submit 提交经验 ```README.mdupstream ↗
Contents are not inlined in the catalog. Open the upstream file.
Plain-text copy, no login: /skill-bundles/ianyang88-openclaw-whoknowsit-skill-skill.md