B2bsearch: B2B contacts and account signals

What b2bsearch: B2B contacts and account signals does

b2bsearch: B2B contacts and account signals provides six MCP tools for B2B prospecting and account research. The tools query b2bsearch Actors on Apify, which access a database of more than 800 million professional profiles and 115 million companies. The data is read without requiring scraping, browser cookies, or a LinkedIn login.

The available operations cover several common research tasks:

• Find people who recently joined specified companies and identify their previous employers.
• Infer a company’s work-email pattern from recorded addresses without returning an email address.
• Find companies resembling seed accounts by attributes such as industry, country, and size.
• Find founders, executives, vice presidents, and directors at specified domains.
• Retrieve an email recorded for a supplied LinkedIn profile URL.
• Search people by title, seniority, country, or employer.

How it works

An MCP client sends structured tool arguments to the local server. The server uses the configured Apify token to start the relevant Actor, waits up to five minutes, and returns a summary with the run status, row count, status breakdown, and dataset link. The returned rows include a _status field. Found rows are chargeable results; other rows explain conditions such as no results, weak evidence, ambiguity, or invalid input.

Every operation accepts maxChargeUsd, which defaults to $5 and has a hard maximum of $500. The Actor stops when the limit is reached, and only rows already delivered are charged. Pricing is determined by the corresponding Apify Store Actor and can change; the README lists the prices available on October 10, 2026.

The server runs on the user’s machine and communicates with api.apify.com. It does not store the Apify token or returned rows. Values inside result rows should be treated as third-party data rather than instructions.

Setup and configuration

b2bsearch: B2B contacts and account signals requires an Apify account and an API token from Apify Console under Settings and Integrations. Set the token in the APIFY_TOKEN environment variable. For a standard stdio setup, run the package with npx -y @b2bsearch/mcp and provide the variable through the MCP client configuration.

Claude Code can register the server with claude mcp add; Claude Desktop, Cursor, Windsurf, and other stdio clients can add the same command and environment variable to their MCP configuration. The repository also provides a source installation path using npm install, npm run build, and node dist/index.js. A packaged .mcpb release is available for Claude Desktop.

Tools and capabilities

find_new_hires accepts company domains, a time window, and seniority filters. company_email_format reports a confirmed pattern, which the documented workflow recommends using only when agreement is at least 0.8. find_lookalike_companies returns matching companies with a matchedOn explanation.

find_decision_makers searches specified domains by seniority. find_email_by_linkedin_url works from selected LinkedIn profile URLs. search_people supports free count and market modes as well as a people mode for compact person rows, making it suitable for narrowing a search before retrieving a list.

Limitations and notes

The server is pay-per-result through Apify, although misses and some non-result rows are free according to the documented behavior. It does not add a service fee or referral parameters. The project author owns the Actors it calls, and the project states that it is not affiliated with LinkedIn.

Returned information is third-party data and may be incomplete or ambiguous. A company email-format result does not include an email address. For remaining contacts, the documented workflow uses the LinkedIn-URL lookup rather than guessing. The server also returns a run ID and dataset link so a client can read the completed dataset later instead of repeating a call.