​Zero to Agentic: Build an API Test Suite With an Agent, From Scratch

​Watching an agent generate tests is one thing. Building the entire setup yourself, seeing it fail, and understanding how it recovers is much more useful.

​In this hands-on workshop, you’ll start with an OpenAPI specification and a terminal and build an agentic API testing setup from the ground up. You’ll move through the full loop: connecting an MCP server, giving the agent your testing standards, generating contract tests, running them, handling a failure, and repairing the test.

​Event Details

• ​Date: 23 September 2026

• ​Time: 12:00 PM ET

• ​Venue: Online

​Session: Zero to Agentic: Build an API Test Suite With an Agent, From Scratch

​Speaker: Somu Suryanarayanan

​The workshop begins with an OpenAPI specification and a real endpoint.

​Somu will show how to connect an MCP server and create a skill file that captures assertion and error-handling standards, then use that setup to generate and execute a contract test suite.

​You’ll also see what happens when the agent encounters a 422 response. Rather than stopping at the failure, the session will show how the agent interprets what went wrong and corrects its own test without human intervention.

​The workshop then moves into AI-assisted CI failure triage, showing how an agent can analyse a failed CI run, identify the root cause, and explain the failure in language that non-technical stakeholders can understand.

​The session will also cover the costs, limitations, and situations where this approach should not be used.

​What You’ll Take Away

• ​A practical mental model of MCP that can transfer across tooling contexts

• ​A skill file encoding assertion and error-handling standards

• ​A contract test suite generated from an OpenAPI specification

• ​First-hand experience of an agent diagnosing and repairing a failing test

• ​A repeatable pattern for AI-assisted CI failure triage

• ​A clearer understanding of the costs and limitations of agentic testing

​How This Huddle Solves Your Problem

​If most agentic testing examples you have seen stop at test generation, this workshop goes further.

​You’ll see how the pieces fit together from the beginning: how testing standards are provided to the agent, how tests are generated from an API specification, how failures are interpreted, and how the same setup can support CI triage.

​The session is also designed to make the boundaries clear. You will see not only what the agent can do, but where this approach becomes costly, unreliable, or inappropriate.

​Why Show Up on 23 September?

​This workshop is for people who want to understand the mechanics of agentic testing rather than only see the finished result.

​You’ll be able to follow the complete workflow, ask questions about the decisions being made, and see how MCP, skills, test generation, failure handling, and CI analysis connect in one practical setup.

​Just as importantly, the session will include the limitations and trade-offs, so you can judge where this approach may or may not fit your own testing workflow.

​Become a Speaker at The Test Tribe Events

​Wish to share your insights with the community at a future event?

​Submit your talk idea here, and if it resonates with the perspective we are trying to build at our events, our team will connect with you.

​About The Test Tribe

​The Test Tribe is the world's largest software testing community, empowering testers globally since 2018. With 925+ events and 170K+ members across 130+ countries, we bring software professionals together to learn, grow, and connect through meetups, conferences, cohorts, masterclasses, and more.

​By RSVPing, you agree to our Terms & Conditions and Privacy Policy and consent to be contacted by The Test Tribe and our event partners.