Govern agents without slowing them.
Turn on spec-based development and every coding agent on the team works inside the same rules: no edit until it declares the story it is implementing, no commit without a Testmaze-Ref, and CI re-checks the range. The traceability matrix shows requirement → acceptance test → work item → commit → verdict → release for every story, exportable for the auditor.
What goes wrong today
No way to mandate a spec
Rules files and plan modes are suggestions. Nothing stops an agent from editing code that no story asked for.
No independent signal
The agent says tests pass. The tests, the results and the verdict all come from the same place that wrote the code.
Tooling fragmentation
Every agent on the team runs tests differently. There is no one surface to hold them all to the same standard.
What changes
No story, no edit
spec.status · spec.start_work · PreToolUse hookWith the policy on, the Claude Code hook blocks file edits until the agent has declared a reviewed story for the branch. The server rejects stories with no acceptance cases or that are not Ready.
No reference, no commit
commit-msg hook · spec check-rangeEvery commit carries a Testmaze-Ref trailer naming the story or case; the git hook adds and checks it, and the CI range check fails the build on an untraced commit.
The chain, on one page
Traceability · spec.traceRequirement → acceptance test → work item → commit → verdict → defects → release, one row per story, with CSV export. The audit trail ISO 9001 / 12207-style processes ask for.
Independent verification by design
testmaze-verifier · recordedByThe verifier runs as a separate subagent and every recorded result says who recorded it. Verdicts come from fixed rules, pinned to the commit.
Built for your perimeter
tmt_* tokens · BYOK · TESTMAZE_MCP_URLPer-workspace tokens isolate every team, bring-your-own AI keys are AES-256-GCM encrypted, and a private deployment keeps the whole loop inside your network.
How it works for engineering leads
Best for: Engineering leads & teams with a policy
Spec-based development is on for this workspace: no story, no code.
You talk to your agent. It picks the tools.
Test Maze for the rest of your team
Ready to give your agent a verifier?
Create a free workspace, generate an MCP token and connect your coding agent in under five minutes.