# mcp-agent

> mcp-agent — sanjanb-mcp-agent. Use this tool when you need to quickly find answers to HR policy questions from your own documents. The mcp-agent retrieves relevant sections from a vector database and generates responses with citations, using AI models like OpenAI or Gemini, or falls back to a basic retrieval-only mode. It is ideal for HR teams seeking to efficiently manage and provide accurate policy information, with integration capabilities via git.

Canonical page: https://skillsregistry.net/skills/sanjanb-mcp-agent  
JSON: https://api.skillsregistry.net/v1/skills/sanjanb-mcp-agent

## Description

A minimalist HR Assistant that answers policy questions from your own documents. It retrieves the most relevant sections from a vector database and generates grounded responses with citations using OpenAI or Gemini. When no LLM is configured, it falls back to a reliable Basic (retrieval‑only) mode.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **Updated:** 2026-09-20

## Source

- **Source listing:** [GitHub](https://github.com/sanjanb/mcp-agent)

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "sanjanb-mcp-agent"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/sanjanb-mcp-agent` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/sanjanb-mcp-agent/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt