
Extract
- Updated April 11, 2026
- Swwyymm/averra-extract-mcp
Averra Extract is a MCP server that fetches URLs and returns cached Markdown with metadata for AI research and pipeline ingestion.
About
dev.averra/extract is an MCP server that converts URLs into clean Markdown plus metadata, with caching suited to agent pipelines. developers use it when they need dependable page text for competitive scans, doc ingestion, or content workflows without maintaining a custom scraper stack. Install the @averra/extract-mcp npm package and wire stdio transport into Claude Code, Cursor, or Codex so tools can fetch once and reuse cached output across threads. It complements manual copy-paste during idea research and becomes a build integration when you pipe extractions into embeddings, eval sets, or growth content drafts. The scope is ingestion, not interactive browsing, so pair it with your own politeness and robots policies for production crawls.
- URL-to-Markdown conversion tuned for AI pipelines
- Response metadata plus caching to avoid repeat fetches
- npm package @averra/extract-mcp with stdio MCP transport
- Version 0.1.5 on the official MCP registry schema
Extract by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
claude mcp add extract -- npx -y @averra/extract-mcpAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| Package | @averra/extract-mcp |
|---|---|
| Transport | STDIO |
| Auth | None |
| Last updated | April 11, 2026 |
| Repository | Swwyymm/averra-extract-mcp ↗ |
What it does
Turn arbitrary URLs into clean Markdown with metadata and caching for RAG, briefs, and agent research loops.
Who is it for?
Best when you're building research agents, newsletters, or RAG prototypes and want one MCP tool instead of bespoke fetch-and-parse scripts.
Skip if: Workflows that need full browser automation, logins, or heavy JavaScript rendering without a separate browser stack.
What you get
Agents receive consistent Markdown and metadata from URLs, with caching so repeated research passes stay fast and cheap.
- Markdown representation of a requested URL
- Structured metadata alongside extracted content
- Cache-backed repeats for the same URL in agent sessions
By the numbers
- Registry and npm version 0.1.5
- stdio transport via @averra/extract-mcp
- GitHub source at Swwyymm/averra-extract-mcp
README.md
Averra Extract MCP Server
MCP (Model Context Protocol) server for Averra Extract — lets AI agents like Claude, Cursor, and ChatGPT convert any webpage into clean, LLM-ready Markdown.
Listed in the official MCP registry as dev.averra/extract. MCP clients that support registry-based install can reference that identifier directly; for clients that need manual configuration, see the install sections below.
What it does
Exposes the Averra Extract API as 5 MCP tools:
| Tool | Description |
|---|---|
averra_extract_url |
Convert any URL to clean Markdown with metadata |
averra_check_usage |
Check monthly quota and remaining requests |
averra_create_api_key |
Create a new API key |
averra_list_api_keys |
List all API keys on the account |
averra_revoke_api_key |
Revoke an API key by ID |
Get an API key
Sign up at averra.dev — the free plan includes 50 requests/month.
Install (Claude Desktop)
Add this to your claude_desktop_config.json:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"averra-extract": {
"command": "npx",
"args": ["-y", "@averra/extract-mcp"],
"env": {
"AVERRA_EXTRACT_API_KEY": "sk_live_your_key_here"
}
}
}
}
Restart Claude Desktop. The 5 tools will appear in your session.
Install (Cursor)
Add to ~/.cursor/mcp.json (or use the Cursor MCP settings UI):
{
"mcpServers": {
"averra-extract": {
"command": "npx",
"args": ["-y", "@averra/extract-mcp"],
"env": {
"AVERRA_EXTRACT_API_KEY": "sk_live_your_key_here"
}
}
}
}
Install (other MCP clients)
Any stdio-based MCP client works. Point it at npx -y @averra/extract-mcp with AVERRA_EXTRACT_API_KEY in the environment.
Configuration
| Env var | Required | Default | Description |
|---|---|---|---|
AVERRA_EXTRACT_API_KEY |
yes | — | Your Extract API key (starts with sk_live_) |
AVERRA_EXTRACT_BASE_URL |
no | https://averra.dev |
Override API host |
TRANSPORT |
no | stdio |
stdio (local) or http (hosted) |
PORT |
no | 3000 |
Port when TRANSPORT=http |
Example prompts (Claude Desktop)
Once installed, try:
- "Extract the content from https://example.com and summarize it"
- "What's my remaining Extract quota this month?"
- "Read https://docs.anthropic.com/en/docs/intro and tell me the key concepts"
- "List my Extract API keys"
License
MIT
Recommended MCP Servers
How it compares
MCP URL ingestion server, not a headless browser skill or site-wide crawler product.
FAQ
Who is Averra Extract for?
Averra Extract is for developers and small teams who use AI agents to research the web and need Markdown-ready page content with metadata and caching.
When should I use Averra Extract?
Use it during idea research, when ingesting docs into a build pipeline, or when refreshing source pages for content and growth workflows.
How do I add Averra Extract to my agent?
Install npm package @averra/extract-mcp (registry version 0.1.5), configure stdio MCP transport in your client, and point tools at the URLs you want converted.