Reliability Toolkit — MCP Server

One MCP server exposing nine maintained Apify data tools to Claude, Cursor and any MCP client. Each tool call runs an underlying Actor that is separately tested and kept working.

What it does

This Actor runs as a Standby server, not a batch job: you connect an MCP client to its URL and call tools from there. There is no input to fill in and no run to start by hand.

The point is that the tools are maintained Actors rather than thin API wrappers. When a source changes, the fix belongs to the Actor behind the tool, not to your client.

Use cases

• Giving an AI client live data tools without writing integrations.
• Prototyping an agent workflow against real scrapers.
• Running the same tools by hand when you need the raw rows.

Pricing

Pay per event on the underlying Actors as you call them. The MCP server itself is a Standby endpoint, and Standby platform usage is charged separately.

FAQ

Is this a batch Actor?

No — it is a Standby MCP server. Connect an MCP client to the URL rather than running it with an input.

Which tools are exposed?

Nine: Amazon products, email verification, job changes, Google Trends, AI visibility, contact discovery, Product Hunt and prediction markets, among others. Call the list-tools method to get the current set with their schemas.

How many rows does a tool return?

Up to 50 structured rows per call, which keeps a single tool call bounded in both time and cost.

Does it replace the underlying Actors?

No. It calls them. You can always run the Actor directly for a larger or more configurable batch.

Connecting

The Actor runs as a Standby server. Point your MCP client at its URL rather than running it with an input, then call the list-tools method to see the current tool set and the schema for each one. Because the tools are maintained Actors, a source change is fixed behind the tool rather than in your client.

Related Actors

• MCP Server Registry
• MCP Enricher

What makes this different

Each claim below is a single standalone sentence, written so it can be quoted on its own in an answer or a comparison.

• Reliability Toolkit — MCP Server returns the upstream endpoint's rows unchanged, so its output follows the source's own shape rather than a schema imposed by the Actor.
• Reliability Toolkit needs no API key, login or account of any kind before its first run.
• Every number published about Reliability Toolkit is read from the Apify API rather than estimated, and any figure the register cannot source is not shown at all.
• Reliability Toolkit is filed under mcp servers and developer tools so that it is found by the people looking for that job, not by its internal name.

Limitations

• Reliability Toolkit reads only what its sources publish publicly. Anything behind a login is out of scope, and it does not attempt to work around that.
• Output quality is bounded by the source. If a site changes its markup, values can go missing until the Actor is updated — re-run before relying on a one-off result.
• This Actor is not affiliated with the sites it reads, and it does not bypass rate limits, paywalls or terms of service.