mcp-company-change-event-feed

Purpose

Company Change-Event Feed MCP Server is used to monitor changes in company domains across hiring, tech stack, funding, company attributes, and social media presence, diff the current state against the last run's results, and return only what changed in the form of typed change events. It is aimed at RevOps teams, outbound agencies, Clay users, and AI agents that need incremental intelligence on target accounts, with output as flat, Clay-ready JSON, one row per company.

Main capabilities

This server provides AI clients with one tool:get_company_changes. It monitors changes in company domains since the last run across hiring, tech stack, funding, company attributes, and social media. The first run records a clean baseline (no changes); every subsequent run returns only deltas, presented as typed change events, each event carrying severity, confidence, old and new values, and the immutable Actor ID of the source that produced the event.

All work runs on Apify. This package is a thin client that routes tool calls to the actor and returns results.

How to use

Requires Node.js 18 or higher, and an Apify account with an API token.

Add this configuration to the Claude Desktop config. Get a token at https://console.apify.com/account/integrations , paste it in, then restart Claude Desktop. The tool is then ready to use.

Example prompts:

• "What changed at stripe.com since the last check across hiring, tech stack, and firmographics?"
• "Monitor hubspot.com and tell me only the high-severity changes."
• "Run a baseline for deel.com (company name Deel), then check it next week for deltas."
• "Did anything change in the GTM signals for notion.so?"

Tool input:

• domain(string, required unless domains is set): the company domain to monitor, without https or www, e.g. stripe.com.
• domains(array, optional): batch mode, monitor multiple company domains in one run, with one output row per domain. When set and non-empty, it overrides domain。
• company_name(string, optional): company name hint, used when the domain does not match the brand name, e.g. Deel for deel.com.
• sources(array, optional): sources to monitor:hiring、tech_stack、firmographic、funding、social. If omitted, all five are monitored. Unselected sources do not run and are not charged, so this is the biggest lever for controlling cost.
• min_severity(string, optional):low(default, all),medium(filter out minor noise),high(only funding, acquisitions, executive changes, CRM changes, and major hiring expansion).
• previous_snapshot(object, optional): a snapshot object returned by a previous run. If provided, it will be used as the baseline instead of the stored snapshot, so you can keep incremental state outside Apify. Ignored in batch mode.
• sub_actor_timeout_secs(integer, optional): timeout per source in seconds, 5 to 300, default 90. Sources run in parallel, so total wall-clock time is roughly equal to the slowest source. Lower it to keep quick tests short.

sources and sub_actor_timeout_secs are the two controls for the duration and cost of a single call: each selected source is one sub-actor run.

Output

Output is one row per company:company_domain、company_name、run_date、is_baseline、total_changes、has_high_severity、latest_change_date, an changes array composed of typed change events, a source_status object (ok / degraded / skipped / not_selected per source), and one carrying the current state for comparison on the next run snapshot object. Each change event contains event_type、severity、confidence、old_value、new_value and the immutable Actor ID of the source_actor.

The funding source runs the Funding and Press Signal Scanner, so funding, executive changes, product launches, and acquisition events are reported like any other source.

Limitations

The tool starts an actor run and polls until completion, so long-running runs will not be cut off at 300 seconds. Runs are allowed 1,800 seconds. If it is still in progress two minutes after that, the call stops waiting and returns the run ID and its link in the Apify Console, where the results will land once complete. Unsuccessful runs are returned as errors with their run ID and status.

Requires Node.js 18 or higher, and an Apify account with an API token.

For the full input and output reference, pricing, and run history, see the Company Change-Event Feed actor on the Apify Store:

https://apify.com/mambalabs/company-change-event-feed

This server is part of the Mamba Labs GTM Suite, a collection of 54 MCP servers for go-to-market data, each backed by a dedicated Apify actor. All servers are listed at https://mambabuilt.com。

Built by Mamba Labs | npm | Apify Store

Built by Mamba Labs. https://apify.com/mambalabs