# PyTorch Documentation Search

> Use this tool when you need to quickly find relevant PyTorch APIs, examples, or error messages in the documentation. It solves problems of inefficient manual searches and provides accurate code-aware results through semantic search capabilities. The tool accepts search queries as input and returns ranked results with prioritized code snippets or conceptual explanations as output, making it ideal for use cases involving implementation queries or theoretical questions.

Canonical page: https://skillsregistry.net/skills/seanmichaelmcgee-pytorch-docs  
JSON: https://api.skillsregistry.net/v1/skills/seanmichaelmcgee-pytorch-docs

## Description

PyTorch Documentation Search Tool provides semantic search capabilities over PyTorch documentation with code-aware results, enabling Claude Code to find relevant APIs, examples, and error messages. Built with a modular architecture supporting both STDIO and SSE transports, it uses vector embeddings through OpenAI's API to deliver high-quality search results with intelligent ranking that prioritizes code snippets for implementation queries and conceptual explanations for theoretical questions. The tool integrates seamlessly with Claude CLI through multiple deployment options including direct STDIO integration, HTTP/SSE for shared servers, and UVX packaging for distribution.

## Trust

- **Trust score (0–1):** 0.51
- **Verification tier:** scanned
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** devops-ci
- **Updated:** 2026-09-28

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/seanmichaelmcgee-pytorch-docs)
- **Repository:** <https://github.com/seanmichaelmcgee/pytorch-docs-refactored>

## 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": "seanmichaelmcgee-pytorch-docs"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/seanmichaelmcgee-pytorch-docs` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/seanmichaelmcgee-pytorch-docs/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