# Search Fragments

> Use this tool when you need to resolve incomplete or half-remembered search queries, and require a reliable output that avoids confident but incorrect responses. It takes incomplete search queries as input and returns a set of possible matching fragments as output. Ideal for use cases where accuracy is crucial, and incorrect information could be misleading.

Canonical page: https://skillsregistry.net/skills/com-searchfragments-search-fragments  
JSON: https://api.skillsregistry.net/v1/skills/com-searchfragments-search-fragments

## Description

For queries a model can't confidently place — resolves or declines. Never a confident-wrong.

## Trust

- **Trust score (0–1):** 0.70
- **Verification tier:** unverified

## Facts

- **Version:** 0.1.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** search
- **Updated:** 2026-07-16

## Source

- **Source listing:** [MCP Registry](https://registry.modelcontextprotocol.io/v0/servers/com.searchfragments%2Fsearch-fragments)

## Use it

MCP endpoint published by the skill: `https://searchfragments.com/api/mcp`

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": "com-searchfragments-search-fragments"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/com-searchfragments-search-fragments` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/com-searchfragments-search-fragments/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