# retrieval

> retrieval — hcarrillo001-retrieval-mcp. Use this tool when you need to evaluate the accuracy and reliability of AI-generated outputs, solving problems of misleading or irrelevant information. It scores outputs for faithfulness, relevancy, and hallucination, accepting custom metrics and golden sets as inputs and providing run history and dashboards as outputs. Ideal for use in MCP client applications where trustworthy AI outputs are crucial.

Canonical page: https://skillsregistry.net/skills/hcarrillo001-retrieval-mcp  
JSON: https://api.skillsregistry.net/v1/skills/hcarrillo001-retrieval-mcp

## Description

Scores AI outputs for faithfulness, relevancy, and hallucination inside any MCP client, with custom metrics, golden sets, and run history with dashboards.

## Trust

- **Trust score (0–1):** 0.68
- **Verification tier:** scanned
- **Last scanned:** 2026-09-04

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** data-analytics
- **Updated:** 2026-09-04

## Source

- **Source listing:** [Glama](https://glama.ai/mcp/servers/bshh1vef8r)
- **Repository:** <https://github.com/hcarrillo001/retrieval-mcp>

## 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": "hcarrillo001-retrieval-mcp"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/hcarrillo001-retrieval-mcp` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/hcarrillo001-retrieval-mcp/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