---
title: "Docs crawler MCP server"
canonical: "https://agenticfeats.com/c/docs-crawler-mcp-server-claude-code/"
agents: ["Claude Code"]
evidence_grade: "A"
outcome: "completed"
job_type: "Coding"
roles: ["Developer"]
source_url: "https://dev.to/sizzlebop/i-gave-my-ai-agents-their-own-documentation-crawler-and-pulled-60-pages-of-clean-markdown-in-49-2cl7"
evidence_url: "https://github.com/pinkpixel-dev/pulpie-mcp"
source_date: "2026-10-05"
published: "2026-10-13"
updated: "2026-10-13"
---

> From Agentic Feats: sourced examples of AI agents at work. Map of the site for agents: https://agenticfeats.com/llms.txt

# Docs crawler MCP server

Claude used the author's pulpie-mcp server to crawl a documentation site and save every page locally as clean Markdown files.

Result: Claude pulled 60 documentation pages into local Markdown files in about 49 seconds, including roughly 10 seconds of backend cold start.

## How it was set up

pulpie-mcp is an MCP server with four tools (fetch_markdown, save_markdown, crawl_docs, list_library). It runs the local Pulpie 210M extraction model on a GPU backend.

## Worth knowing

The 49-second figure includes a one-time backend start and model load. The author's own tool, tested on one 60-page docs site.

## Prompt

```text
Pull the [library] docs into this project's [folder] folder.
```

## Evidence note

Open-source repo with README; author reports timing table for the 60-page crawl

## Sources

- Original: https://dev.to/sizzlebop/i-gave-my-ai-agents-their-own-documentation-crawler-and-pulled-60-pages-of-clean-markdown-in-49-2cl7
- Evidence: https://github.com/pinkpixel-dev/pulpie-mcp

## The receipt

- Agent: Claude Code
- Evidence: Shown (grade A, artifact linked)
- Outcome: Completed
- Job type: Coding
- Roles: Developer
- Source: Blog
- Published: 2026-10-05
- Added: 2026-10-13