# fetch-docs-haiku45

fetch-docs-haiku45 retrieves online documentation using the Haiku model and saves it as markdown without cuts.

- 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-haiku45.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-haiku45

## What it does

This subagent fetches web documentation and writes the exact text into a target markdown file. Running on the `haiku` model, it is tailored for extracting technical materials and benchmarking document retrieval workflows.

For pulling content, the agent relies on `mcp__firecrawl-mcp__firecrawl_scrape`, falling back to `WebFetch` when needed. It does not summarize content or drop technical sections: code snippets, formatting, and examples remain untouched. The full document is written via the `Write` tool, after which the agent reports the final path.

Without an active Firecrawl MCP configuration, the agent must rely on standard `WebFetch` requests.

## Install

### Claude Code (native platform)

**Manually: put the files in place**
- File: `.claude/agents/fetch-docs-haiku45.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-haiku45.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-haiku45.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-haiku45.md

## When to use it
- pull raw documentation from web pages into markdown files
- preserve code snippets and formatting without model summarization
- benchmark documentation retrieval workflows with Haiku

## Example prompts
- Fetch the online guide from the provided link and write it directly to `docs/reference.md`
- Grab the technical specification page and save every section into `output/api.md` without summarizing

## Similar
- [firecrawl-knowledge-ingest](https://affpapa.org/en/nevibecode/skills/firecrawl-knowledge-ingest.md) — skill · 31600 installs
- [verify-docs-findable](https://affpapa.org/en/nevibecode/skills/verify-docs-findable.md) — skill · 340 installs
- [firecrawl-crawl](https://affpapa.org/en/nevibecode/skills/firecrawl-crawl.md) — skill · 81400 installs
- [fetch-tweet](https://affpapa.org/en/nevibecode/skills/fetch-tweet.md) — skill · 901 installs
- [research-external](https://affpapa.org/en/nevibecode/skills/research-external.md) — skill · 505 installs
- [social-fetch](https://affpapa.org/en/nevibecode/skills/social-fetch.md) — skill · 531 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
