MCP Gtm Suite

What mambalabsdev/mcp-gtm-suite does

mambalabsdev/mcp-gtm-suite puts 21 GTM-focused tools behind one MCP endpoint. The tools can resolve a company identity, enrich firmographics and social profiles, inspect hiring and job-board activity, detect GTM technology, find funding and press events, and calculate composite or ICP-specific scores. Additional tools analyze AI adoption, outbound infrastructure, editorial output, agent accessibility, public event participation, legal entities, public awards, and LinkedIn activity.

Results are generally returned as flat JSON rows designed for downstream use in Clay-style workflows. Several tools include confidence scores, evidence, provenance, completeness fields, change events, or match details so an agent can distinguish a result from the supporting signals.

How it works

The package acts as a thin MCP client for Mamba Labs actors on Apify. An MCP tool call is routed to the corresponding actor, which performs the research and returns its result. Most operations are read-only and consume Apify credits per call.

Inputs vary by tool. Company-oriented operations commonly accept a domain, company name, or LinkedIn URL. score_icp_fit accepts an ICP template, scoring configuration, or plain-language description; a plain-language description can use an LLM key. aggregate_gtm_signals combines hiring and technology checks and can include additional sources. get_company_changes compares a supplied prior snapshot, while push_leads_to_sequencer maps lead fields and can apply ICP and duplicate filters.

Setup and configuration

The runtime requirement is Node.js 18 or newer, an Apify account, and an APIFY_TOKEN. The package can be launched with npx and configured as an MCP server in Claude Desktop:

{
  "mcpServers": {
    "mamba-gtm-suite": {
      "command": "npx",
      "args": ["-y", "@mambalabsdev/mcp-gtm-suite"],
      "env": {
        "APIFY_TOKEN": "your-apify-token"
      }
    }
  }
}

Restart the client after adding the configuration. Apify usage is charged per actor call. ICP scoring can accept an llm_api_key input for language-model-based ICP descriptions, and contact classification can use an LLM fallback through an LLM_API_KEY configured on the relevant actor. push_leads_to_sequencer also requires the target campaign to already exist and uses the destination vendor's API credentials through its actor configuration.

Tools and capabilities

• Resolve company names, domains, LinkedIn URLs, legal entities, and social profiles.
• Enrich employee bands, industries, headquarters, founding years, revenue estimates, logos, and descriptions.
• Detect hiring, job-board, GTM stack, funding, AI, outbound, publication, event, and public-award signals.
• Score ICP fit or combine GTM signals into a composite score with recommended actions.
• Audit site accessibility for AI agents, including crawler policies, structured data, and machine-readable endpoints.
• Classify contacts by department and seniority using deterministic rules.
• Push qualified leads to existing Instantly or Smartlead campaigns, with dry_run support.

Limitations and notes

Apify credits are consumed by the research tools, and public-source coverage varies by site and source. LinkedIn capture is limited to information visible without login, and degraded results should be checked before treating an absence as definitive. Publication counts may be estimates; use the returned reliability and estimation fields before quoting them. Legal-entity matching defaults to strict normalized-name agreement, so unresolved results can be preferable to an uncertain match. Sequencer delivery writes to a campaign unless dry_run is enabled, and the tool does not create campaigns or sequence copy.