# Search Fragments

Tracks down movies, books, songs, and research papers from vague clues. Helps AI agents avoid fabricating answers when details are fuzzy.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: Searchfragments
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/Searchfragments/search-fragments-mcp
- Updated: 2026-10-13
- HTML: https://affpapa.org/en/nevibecode/mcp/search-fragments

## What it does

AI agents often invent confident falsehoods when asked about half-remembered topics. Search Fragments fixes this by enforcing three strict resolution outcomes. The server delivers a verified answer with evidence, a ranked shortlist, or an explicit refusal to guess. Standard search handles simple facts faster.

The service runs as a hosted endpoint at `https://searchfragments.com/api/mcp` without local package downloads. It exposes three MCP tools to the agent: `resolve_fragment` for uncovering sources from vague memories, `verify_claim` for fact-checking specific statements, and `submit_resolution_feedback` for scoring results.

Basic access is free. It requires no account. When authentication is needed, OAuth 2.0 with PKCE is supported out of the box.

## Requirements
- Remote server: `https://searchfragments.com/api/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http search-fragments https://searchfragments.com/api/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add search-fragments --url https://searchfragments.com/api/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "search-fragments": {
            "url": "https://searchfragments.com/api/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "search-fragments": {
            "type": "http",
            "url": "https://searchfragments.com/api/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http search-fragments https://searchfragments.com/api/mcp
```

## When to use it
- Identify a forgotten film or book from fragmentary plot memories
- Verify a dubious claim without triggering AI hallucinations
- Locate an obscure academic paper from vague methodology details
- Return an honest refusal when available clues do not point to a single source

## Example prompts
- Find that 1990s movie where a traveler gets trapped at a desert gas station.
- Verify whether a 2018 experiment tested caffeine effects on mouse memory.
- Locate a novel where the protagonist collects antique keys in a coastal town.

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 392262
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33989
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81163
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119231
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47577
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43902

---
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