A quality gate your coding agent cannot grade itself
AI agents now write most of the code and then mark their own homework. Test Maze is the second pair of eyes: it holds the features, user stories and test cases, records every run against the exact commit, and returns a pass or fail decided by fixed rules — not by a model.
Why we built it
Coding agents are fast, and they are convincing. Ask one whether the change is done and it will tell you yes, usually with a tidy summary of tests it believes it ran. There is no independent record: the tests, the results and the verdict all come from the same place that wrote the code. Three things go wrong, every time — the bar drifts from one session to the next, nothing is comparable across a team, and flaky work hides behind a confident paragraph.
We split the job. Your agent keeps the source tree and runs the tests. Test Maze keeps the test artefacts, stores each result against its git sha, and grades the change deterministically. When it says pass, you can point at the commit, the cases and the run that made it pass.
How the loop works
Everything happens over the Model Context Protocol (MCP), so it works with Claude Code, Cursor, Cline, Gemini CLI, Codex CLI and any other MCP client. You talk to your agent in plain English; the agent calls the tools.
project.initialize— point the workspace at the project your agent is working in.feature.implement— turn a feature description into user stories and acceptance criteria.feature.verify— generate test cases that cover those criteria.testrun.createthentestrun.record_results— your agent runs the tests and reports what happened, with the git sha.pdlc.verify— the deterministic verdict, plus a next step: ship, repair the code, repair the test or add coverage.
Around that spine sit exploratory testing with the agent's own browser, coverage gap analysis, frozen regression baselines, acceptance-criterion waivers and code-quality checks — 55 tools in all.
What we hold to
The verdict is computed, never guessed
pdlc.verify grades a change with fixed rules over the recorded test results: pass when at least one case passed and none failed, fail when a case failed or a frozen regression baseline is missing or failing. The same results always give the same verdict. AI may add a hint about what to fix — it never decides.
We never read your repository
Test Maze does not clone, scan or index your code. It only receives what your agent sends in a tool call: features and user stories, test cases, run results, git identifiers, optional screenshots, and a code snippet when you explicitly ask for a code-quality check.
Every result is tied to a commit
Each verifier test run records the git sha, the branch and whether the working tree was clean. A green run means a specific commit was green — not "it worked on someone’s machine at some point".
Your agent stays in charge
The coding agent owns the source tree and runs the tests. Test Maze owns the test artefacts and the grading. Neither side has to trust the other’s summary of what happened.
Who it is for
Teams shipping with AI agents who still want to answer "how do you know it works?" — QA engineers who want the test estate in one place instead of scattered across chat logs, automation engineers who want runs tied to commits, and engineering leads who want one quality bar that every agent session is measured against.
Who builds it
Test Maze is built by Ceegees, a software engineering company that has spent years building and testing products for other teams. We wanted the tool for our own agent-assisted work first.
Questions, feedback or a pilot for your team? Talk to us — or read the FAQs and pricing.
Ready to give your agent a verifier?
Create a free workspace, generate an MCP token and connect your coding agent in under five minutes.