# GroundLens

Fast guardrails MCP server that validates model answers against source passages without external LLM calls.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 7
- Author: groundlens-dev
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/groundlens-dev/groundlens
- Updated: 2026-10-19
- HTML: https://affpapa.org/en/nevibecode/mcp/groundlens

## What it does

GroundLens checks generated figures and subjects against reference passages. The verification runs in roughly 10 ms via a deterministic Rust engine. It requires no network requests or external model calls for primary checks.

Every run assigns a status of `PASS`, `REVIEW`, or `FAIL` and writes a signed record to disk. Wrong outputs stop here, keeping non-viable answers away from an LLM judge. The server integrates with Langfuse and OpenTelemetry pipelines. Running it requires `uvx`.

## Requirements
- Needs: Python and uv (commands run via uvx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add groundlens -- uvx groundlens
```
uvx runs the command named after the package; if the README names another one, use uvx --from groundlens <command>

### Codex

**Platform CLI**
```bash
codex mcp add groundlens -- uvx groundlens
```
uvx runs the command named after the package; if the README names another one, use uvx --from groundlens <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "groundlens": {
            "command": "uvx",
            "args": [
                "groundlens"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from groundlens <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "groundlens": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "groundlens"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from groundlens <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add groundlens uvx groundlens
```
uvx runs the command named after the package; if the README names another one, use uvx --from groundlens <command>

## When to use it
- verify numerical claims against reference text
- catch unsupported statements before querying an LLM judge
- record cryptographically signed logs for validation audits
- stream guardrail telemetry into Langfuse or Grafana

## Example prompts
- Verify this financial summary against the source text using GroundLens
- Check if the numbers in the answer match the provided reference passages
- Run a GroundLens validation and save the decision record

## Similar
- [MCP Toolbox for Databases](https://affpapa.org/en/nevibecode/mcp/googleapis-mcp-toolbox-for-databases.md) — MCP server · ★ 16730
- [labelhead-artist-momentum](https://affpapa.org/en/nevibecode/mcp/labelhead-artist-momentum.md) — MCP server · ★ 107701
- [Agent Skills Search Server](https://affpapa.org/en/nevibecode/mcp/agent-skills-search-server.md) — MCP server · ★ 26673
- [Skyvern](https://affpapa.org/en/nevibecode/mcp/skyvern.md) — MCP server · ★ 23283
- [Nhost](https://affpapa.org/en/nevibecode/mcp/nhost.md) — MCP server · ★ 9369
- [PraisonAI](https://affpapa.org/en/nevibecode/mcp/praisonai.md) — MCP server · ★ 9298

---
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