pline.ai
Scrape, crawl, map and search the web from your MCP client with a pline.ai key
- Platforms
- CLI, API, macOS
- Runs
- Locally
- Version
- 0.3.12
- Listing updated
- 4 days ago
Similar MCP servers
About pline.ai
pline.ai is an MCP server for web search and scraping. It runs locally on your machine and is started by your AI client.
This is a small local launcher that sits between an MCP client and the hosted pline.ai API. The scraping backend stays on pline.ai's servers, so there is no orchestrator, browser fetcher, Redis or Temporal to run locally. Each user brings their own API key, sent to the API as an x-api-key header.
It installs through npx as the pline-api package (Node.js 18+, with prebuilt binaries for macOS, Linux and Windows), as a Docker image on GHCR, or from source with Cargo. A scrape remembers its session per host, so a second scrape of the same host reuses it unless a new session is requested. Over stdio, results can be saved to a local folder; otherwise output is clipped at 60,000 characters by default and the truncated fields are listed.
A remote endpoint with Pline sign-in is described as an alternative to pasting keys, and the repository also installs as a Claude Code plugin.
Install
Run in a terminal
claude mcp add pline-api --env PLINE_API_KEY='<PLINE_API_KEY>' -- npx -y pline-apiAdd to claude_desktop_config.json
{
"mcpServers": {
"pline-api": {
"command": "npx",
"args": [
"-y",
"pline-api"
],
"env": {
"PLINE_API_KEY": "<PLINE_API_KEY>"
}
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"pline-api": {
"command": "npx",
"args": [
"-y",
"pline-api"
],
"env": {
"PLINE_API_KEY": "<PLINE_API_KEY>"
}
}
}
}Add to .vscode/mcp.json
{
"servers": {
"pline-api": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"pline-api"
],
"env": {
"PLINE_API_KEY": "<PLINE_API_KEY>"
}
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"pline-api": {
"command": "npx",
"args": [
"-y",
"pline-api"
],
"env": {
"PLINE_API_KEY": "<PLINE_API_KEY>"
}
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"pline-api": {
"command": "npx",
"args": [
"-y",
"pline-api"
],
"env": {
"PLINE_API_KEY": "<PLINE_API_KEY>"
}
}
}
}Using several servers? Build one config file with the MCP config generator.
Configuration
PLINE_BASE_URLpline.ai API base URL. Optional; defaults to https://apix.pline.ai/v1.PLINE_API_KEYYour own pline.ai API key. Sent to the API as the x-api-key header.RequiredPLINE_BASE_URLpline.ai API base URL. Optional; defaults to https://apix.pline.ai/v1.PLINE_API_KEYYour own pline.ai API key. Sent to the API as the x-api-key header.Required
Tools this server provides (10)
Taken from the project documentation; the running server may differ.
- scrape
- Scrapes a page through the hosted API.
- batch_scrape_start
- Starts a batch scrape.
- batch_scrape_status
- Checks a batch scrape.
- batch_scrape_cancel
- Cancels a batch scrape.
- crawl_start
- Starts a crawl.
- crawl_status
- Checks a crawl.
- crawl_cancel
- Cancels a crawl.
- map_site
- Discovers URLs on a site.
- search
- Searches the web.
- sessions_list
- Lists remembered host-to-session mappings for the caller.
What it does
- Single-page scrape
- Fetch one page, with session reuse per host.
- Batch scraping
- Start, check and cancel batch scrape jobs.
- Site crawling
- Start, check and cancel crawls of a site.
- URL mapping
- Discover the URLs on a site.
- Web search
- Run searches through the hosted API.
- Local file saving
- Over stdio, save outputs or download job artifacts to a local folder.
Good for
- Pull page content into an assistant's context
- Crawl a documentation site
- List a site's URLs before scraping
- Run large scraping jobs and check their status from chat
Pricing
Pricing unknown
Needs a pline.ai API key; plan prices not published
pline.ai FAQ
Do I need an API key?
Yes, your own pline.ai key, set as PLINE_API_KEY.
What does it need to run?
Node.js 18 or later for the npx route, Docker for the image, or Rust to build from source.
Does a scrape run on my machine?
No. The launcher calls the hosted pline.ai API.
How long can output be?
60,000 characters by default unless saved to a local folder.
How do I add pline.ai to Claude Code?
Run this command in your terminal: claude mcp add pline-api --env PLINE_API_KEY='<PLINE_API_KEY>' -- npx -y pline-api
What are the best alternatives to pline.ai?
The closest alternatives to pline.ai are Scrapling MCP Server, Tavily MCP, Oxylabs MCP, apinoa and MTScraper.
Where this comes from
- Description: written from the project's README.
- Install steps: from the official MCP Registry entry.
