Controlled agent integration

MCP tools for AI agents

The MCP integration exposes approved project-scoped testing tools to an AI agent so it can discover capabilities, create or run assets, and return structured evidence without receiving unrestricted platform access.

What you can do

Connect an MCP-capable agent through the documented endpoint and project credentials.
Discover approved tools for specifications, flows, Craftsman, Sentinel, and evidence.
Keep every operation constrained by identity, project, scope, and subscription capability.

Starts with

A project-scoped integration identity, approved scopes, and an MCP-compatible client.
Natural-language tasks that map to the tools exposed to that identity.

Produces

Created or executed testing assets and structured evidence returned to the agent.
Auditable operations limited to the project and permissions granted by the user.

Use it when

Use MCP when an AI assistant should operate SpecTestEngine tools instead of only discussing test design.
Use the public SDK when a deterministic application integration is a better fit than an agent.

Let a user-authorized AI agent create and run tests for you.

MCP is the bridge between SpecTestEngine and external AI agents. A user can give an agent a scoped API key, ask it to connect to the MCP WebSocket endpoint, and let it use the documented tools to create specifications, run checks, inspect evidence, and return a clear testing report.

ConnectionWebSocket + API key

Primary usersDevelopers, QA, AI agents

Main outcomeTests created, run, and explained

Create test assets from a user request

An authorized agent can read the MCP tool descriptions, create REST, SOAP, GraphQL, or flow assets, and keep the user in the loop instead of asking them to click through every form manually.

Run specs and flows

The agent can execute generated or existing tests, inspect run evidence, compare status/body/header mismatches, and return a clear summary of what passed or failed.

Use product knowledge safely

MCP describes the available actions and expected payloads. This gives the agent a stable contract for working in SpecTestEngine without guessing UI structure.

How a user asks an agent to connect

1Give the agent an API keyThe user creates or provides a scoped key for the project. The agent never needs the user's browser session.

2Connect over WebSocketThe agent opens the SpecTestEngine MCP WebSocket endpoint and authenticates with that key.

3Discover toolsThe agent reads the available tool descriptions, input schemas, and project context before creating or running anything.

4Create, run, and reportThe agent creates specs or flows, runs them, checks the evidence, and sends the user a human-readable result.

Example request to an AI agent

Connect to my SpecTestEngine MCP endpoint with this project API key. Create REST specs for the checkout API, run them, inspect failures, regenerate unstable assertions when the live response proves the contract changed, and give me a short report with links to the created assets.

Project-scoped access

The API key should be scoped to the project and permissions the user wants the agent to use. The agent should not receive personal browser credentials.

Tool descriptions guide the agent

The MCP server describes available operations and payloads, so the agent can act through stable product capabilities instead of relying on brittle screen automation.

Human-readable output

The agent should finish by telling the user what it created, what it ran, which evidence changed, and what needs attention.