# ia-qa mcp llm and RAG testing - Dev/QA toolbox

The ia-qa mcp llm and RAG testing - Dev/QA toolbox MCP server connects an AI coding client to tools for checking LLM agents, RAG flows, and everyday Dev/QA data tasks.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 12266 installs
- Author: ia-qa
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/ia-qa/api
- Updated: 2026-04-01
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/ia-qa-mcp-llm-and-rag-testing-devqa-toolbox

## What it does

ia-qa gives an AI agent tools for working with LLM and RAG test data. It can estimate token counts and model cost, split text for retrieval pipelines, pull JSON out of mixed model output, and compare prompt revisions. The toolbox also covers JSON, Base64, URLs, CSV, JWTs, regex, HTML, timestamps, and text statistics.

The service is added as an HTTP MCP endpoint at `https://api-ia-qa.run.tools`. It can be connected to Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and the other listed clients. No API key or account registration is required, so the practical requirement is an MCP configuration in the client.

## Requirements
- Remote server: `https://api-ia-qa.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http api https://api-ia-qa.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add ia-qa-mcp-llm-and-rag-testing-devqa-toolbox --url https://api-ia-qa.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "ia-qa-mcp-llm-and-rag-testing-devqa-toolbox": {
            "url": "https://api-ia-qa.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "ia-qa-mcp-llm-and-rag-testing-devqa-toolbox": {
            "type": "http",
            "url": "https://api-ia-qa.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http ia-qa-mcp-llm-and-rag-testing-devqa-toolbox https://api-ia-qa.run.tools
```

## When to use it
- check whether a prompt fits the model context
- split source text into RAG chunks
- recover JSON from an LLM response
- compare two prompt or config revisions
- estimate API cost from token counts

## Example prompts
- Count the tokens in `prompt.txt` before sending it to the model and estimate the request cost.
- Split this document into overlapping chunks for a RAG pipeline.
- Test this string as JSON and return the formatted result only if it is valid.

## Similar
- [ai-system-testing](https://affpapa.org/en/nevibecode/skills/ai-system-testing.md) — skill · 1400 installs
- [testing-llm](https://affpapa.org/en/nevibecode/skills/testing-llm.md) — skill · 126 installs
- [llm-risk-assess](https://affpapa.org/en/nevibecode/skills/llm-risk-assess.md) — skill · 186 installs
- [llm-testing](https://affpapa.org/en/nevibecode/skills/llm-testing.md) — skill · 621 installs
- [evaluate-rag](https://affpapa.org/en/nevibecode/skills/evaluate-rag.md) — skill · 1100 installs
- [rag-engineer](https://affpapa.org/en/nevibecode/skills/sickn33-rag-engineer.md) — skill · 1000 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
