com.pulsemcp/pulse-fetchweb-automation

pulse-fetch

Verified · yesterday

MCP server for fetching web resources with content extraction and anti-bot bypass.

Install

claude mcp add pulse-fetch -- npx -y @pulsemcp/[email protected]

Requires: FIRECRAWL_API_KEY, BRIGHTDATA_API_KEY, STRATEGY_CONFIG_PATH, OPTIMIZE_FOR, MCP_RESOURCE_STORAGE, MCP_RESOURCE_FILESYSTEM_ROOT, SKIP_HEALTH_CHECKS, LLM_PROVIDER, LLM_API_KEY, LLM_API_BASE_URL, LLM_MODEL

Our take

PulseMCP's flagship 'productionized' fetch server: a single scrape tool that strips pages to clean markdown via Readability, caches results as MCP Resources, and escalates through Firecrawl or BrightData when it hits anti-bot walls — learning which strategy works per URL pattern. Works out of the box with no keys for ordinary pages; the anti-bot bypass and LLM-powered extraction each need their own paid API keys. One of the better-engineered fetch servers, from the team behind the pulsemcp.com directory.

reviewed by hand · 2026-07-05

Something wrong or dead here? Report it

Verification record

last verified: yesterday
github stars: 72
last commit: yesterday
archived: no
license: MIT
npm: @pulsemcp/pulse-fetch
latest version: 0.3.3
in registry since: 2026-06-17

github.com/pulsemcp/mcp-servers