Skip to main content
Glama
lobstrio

lobstr-mcp

Official
by lobstrio
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

Not the docs MCP. docs.lobstr.io/mcp answers "how does lobstr.io work?". This one is "do this in my account": it authenticates as a user and spends real credits.

Quick start

No directory listing or approval needed — this server implements Dynamic Client Registration, so any MCP client can add it by URL and register itself. Authenticating opens Lobstr's consent page in a browser; click Allow and you're back in the client.

Client

Setup

Claude Code

claude mcp add --transport http lobstr https://mcp.lobstr.io/mcp

Claude Desktop, Cursor, Windsurf

Add the block below to the client's MCP config

Claude.ai, ChatGPT

Settings → connectors → add custom connector → paste the URL

VS Code

.vscode/mcp.json, add a server entry pointing at the URL

Codex CLI

Register as a remote MCP server in ~/.codex/config.toml

{
  "mcpServers": {
    "lobstr": {
      "type": "http",
      "url": "https://mcp.lobstr.io/mcp"
    }
  }
}

Exact config keys vary by client and version — check your client's MCP docs if the block above doesn't match. HTTPS is mandatory: the hosted server only works over https://.

Related MCP server: web-scrape

Tools

20 tools. There is no per-scraper tool — new lobstr.io scrapers work with zero code changes. Full behaviour, error codes and the intended call sequence: docs/tools.md, docs/errors.md.

Tool

Does

search_scrapers

Find scrapers matching a query

get_scraper_details

A scraper's input schema, outputs and pricing

list_my_scrapers

The user's saved scraper configurations

get_my_scraper

One saved configuration by id

create_squid

Save a new scraper configuration without running it

add_tasks

Add input rows to a saved configuration

estimate_run

Authoritative cost/time estimate before running

run_scraper

Configure and run a scraper in one call — spends credits

attach_account

Link a connected platform account to a scraper

empty_scraper

Clear a scraper's saved inputs, keep its config

deactivate_scraper

Free a scraper's concurrency slot without deleting it

get_run

Status and progress of a run

list_runs

Recent runs for a scraper

get_results

One page of a run's results

get_results_url

A signed download URL for a run's full results

abort_run

Stop a running job

list_accounts

The user's connected platform accounts

get_account

One connected account by id

check_credits

Credit balance and concurrency slots

whoami

The authenticated user and server version

Self-hosting

git clone https://github.com/lobstrio/lobstr-mcp.git
cd lobstr-mcp
cp .env.example .env    # then edit — see docs/configuration.md
docker compose up -d --build

Full runbook, production hardening and log retention: DEPLOY.md. Environment variables: docs/configuration.md.

Development

uv sync --extra dev
uv run pytest

See CONTRIBUTING.md for the full dev setup and PR flow.

Learn more

lobstr.io: website · docs · scraper store · Python SDK · CLI

Contributing

Issues and pull requests welcome — see CONTRIBUTING.md. Security issues: SECURITY.md, not a public issue. This project follows the Contributor Covenant.

License

Apache License 2.0

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • JobScannerMCP

    singhalpooja9
    F
    license
    Not graded
    quality
    D
    maintenance
    Exposes job-posting scanner tools (search, company health, fit scoring, dry-run outreach) via MCP, enabling offline job search and evaluation from any MCP host.
    -

  • web-scrape

    teslashibe
    A
    license
    Not graded
    quality
    B
    maintenance
    Enables scraping and fetching websites with protection handling like Cloudflare and captchas, via an MCP interface.
    6 npm
    1
    Apache 2.0

  • WebSearchAndCrawl

    bpweatherill
    F
    license
    Not graded
    quality
    F
    maintenance
    Enables authenticated web crawling via Firefox session tokens, regex-based content search, document parsing/downloading, and real-time streaming of results for MCP clients.
    -

  • openactors

    sahilrit
    A
    license
    A
    quality
    B
    maintenance
    Enables MCP clients to run Apify-compatible web scrapers locally via stdio or HTTP, including site crawling, job board retrieval, web search, Google Maps and LinkedIn job lookups, with run management, dataset storage, and a REST API.
    26
    MIT