Agent Knowledge
About
Agent Knowledge is an MCP (Model Context Protocol) server that powers AI Agent job interview simulations with a RAG knowledge base. It integrates FAISS vector search and a Chinese embedding model (BAAI/bge-base-zh-v1.5) to provide semantic retrieval from the hello-agents…
Details
- Author
- Ddhjx-code
- GitHub stars
- 2
- Downloads
- 154
- Categories
- AI
Jump to
- Semantic knowledge retrieval via FAISS vector index (1086 vectors, 768-dim)
- Three MCP tools: search_knowledge, get_interview_questions, get_learning_path
- Uses BAAI/bge-base-zh-v1.5 for Chinese embeddings
- Knowledge sources include 16-chapter tutorial and 90+ external resources
- Zero-infrastructure, single-file persistence for vector store
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Agent KnowledgeCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Install dependencies (pip install -r interview_rag_server/requirements.txt), build the FAISS index by running the provided scripts, then configure the MCP server in .claude/settings.json. Use it through Claude Code’s /interview skill; no direct invocation is needed.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"agent knowledge": {
"interview-rag": {
"command": "uvx",
"args": [
"mcp-server-interview-rag"
]
}
}
}
}McpServers
{
"interview-rag": {
"command": "uvx",
"args": [
"mcp-server-interview-rag"
]
}
}Agent Interview
AI Agent 岗位模拟面试系统 —— 基于 Claude Code Skill + MCP Server + RAG 知识库，为求职者提供沉浸式的 AI Agent 技术面试模拟与评估。
功能
- 简历解析：读取候选人简历（PDF/MD/TXT），分析技术背景和目标岗位
- 多轮面试：覆盖 LLM、Agent、RAG、记忆系统、多智能体、RLHF、VLM、工程实践等方向
- 自适应难度：根据候选人回答质量动态调整问题深度
- RAG 知识库：基于 hello-agents（16章系统教程）、Agent-Learning-Hub（学习路线图 + 90+ 外部资源）构建
- 评估报告：生成包含知识评分、学习方向、改进建议和简历修改建议的完整报告
架构
┌─────────────────────────────────────────────┐
│ Claude Code │
│ │
│ /interview [resume_path] │
│ ↓ │
│ ┌─────────────────────────────────────┐ │
│ │ agent-interview Skill │ │
│ │ 面试流程编排 + 评估 + 报告生成 │ │
│ └──────────────┬──────────────────────┘ │
│ │ MCP tool calls │
│ ┌──────────────▼──────────────────────┐ │
│ │ interview-rag MCP Server │ │
│ │ │ │
│ │ search_knowledge 语义检索知识库 │ │
│ │ get_interview_questions 获取面试题 │ │
│ │ get_learning_path 生成学习路径 │ │
│ │ │ │
│ │ FAISS + bge-base-zh-v1.5 │ │
│ └─────────────────────────────────────┘ │
└─────────────────────────────────────────────┘
快速开始
1. 安装依赖
pip install -r interview_rag_server/requirements.txt
需要的核心依赖：
- fastmcp >= 2.0.0
- faiss-cpu >= 1.7.4
- sentence-transformers >= 2.2.0
2. 构建知识库索引
首次使用需要从源项目构建向量索引：
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
