# fetch-docs-sonnet45

The fetch-docs-sonnet45 subagent retrieves documentation from a URL and writes it to a chosen file. Use it for documentation retrieval benchmarks.

- subagent · `type=agent` · Claude Code, Cursor, Codex · ★ 1573 repo stars
- Author: disler
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/disler/claude-code-hooks-multi-agent-observability/tree/HEAD/.claude/agents/fetch-docs-sonnet45.md
- Updated: 2026-02-08
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/agents/fetch-docs-sonnet45

## What it does

\`fetch-docs-sonnet45` takes a documentation URL and saves the retrieved material to the output path you provide. Its task is to keep the page content intact, including code blocks, examples, and formatting, rather than turning it into a summary.

The subagent has access to `Write`, `Bash`, `Read`, `WebFetch`, and `mcp__firecrawl-mcp__firecrawl_scrape`. Its frontmatter selects the `sonnet` model and describes the agent as using Sonnet 4.5. It gives priority to `mcp__firecrawl-mcp__firecrawl_scrape`, so that MCP tool is the main retrieval route; `WebFetch` is the alternative listed beside it.

## Install

### Claude Code (native platform)

**Manually: put the files in place**
- File: `.claude/agents/fetch-docs-sonnet45.md` — For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

### Codex

**Manually: put the files in place**
- File: `.codex/agents/fetch-docs-sonnet45.toml` — Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

### Cursor

**Manually: put the files in place**
- File: `.cursor/agents/fetch-docs-sonnet45.md` — Cursor also reads .claude/agents/

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/disler/claude-code-hooks-multi-agent-observability/tree/HEAD/.claude/agents/fetch-docs-sonnet45.md

## When to use it
- retrieve documentation from a URL
- save a page as a Markdown file
- keep code examples in the saved content
- benchmark documentation retrieval
- report the output file path

## Example prompts
- Run `fetch-docs-sonnet45` on this URL and save the retrieved documentation to the specified file.
- Use `mcp__firecrawl-mcp__firecrawl_scrape` to fetch the page, then write the result to the requested Markdown path.
- Retrieve the documentation without summarizing it and report where the file was saved.

## Similar
- [pulse-fetch](https://affpapa.org/en/nevibecode/mcp/pulse-fetch.md) — MCP server · ★ 81 repo stars
- [web-fetch-mcp](https://affpapa.org/en/nevibecode/mcp/web-fetch-mcp.md) — MCP server · ★ 9 repo stars
- [Defuddle Fetch](https://affpapa.org/en/nevibecode/mcp/defuddle-fetch.md) — MCP server · ★ 8 repo stars
- [felo-web-fetch](https://affpapa.org/en/nevibecode/skills/felo-web-fetch.md) — skill · 117 installs
- [document-generator-expert](https://affpapa.org/en/nevibecode/agents/document-generator-expert.md) — subagent · ★ 372 repo stars
- [app-docs](https://affpapa.org/en/nevibecode/skills/app-docs.md) — skill · 1100 installs

From the same set: [disler/claude-code-hooks-multi-agent-observability](https://affpapa.org/en/nevibecode/by/disler/claude-code-hooks-multi-agent-observability.md) — 20 cards

---
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
