Copilot Studio Kit (Power CAT)
Microsoft Power CAT's open-source kit: test agents at scale (bulk and automated), manage configuration, and analyze results. Quality-assurance infrastructure for makers and CoE teams.
This is a summary with a canonical link to the original Microsoft resource; up-to-date content lives at the source. — Microsoft Power CAT (GitHub).
Testing one agent by hand is easy; testing dozens of agents consistently and repeatably, release over release, is hard. Power CAT’s (Microsoft Customer Advisory Team) open-source Copilot Studio Kit fills exactly that gap.
Highlights:
- Bulk / automated conversation testing: define test sets, run the agent over hundreds of inputs, compare against expected responses.
- Configuration management: keep agent settings central and traceable.
- Analytics dashboards: report on test results, response quality and regressions.
- Installs into Power Platform as a managed solution; built on Dataverse.
Honest limit: this is a community/Power CAT project — no formal product SLA — but it’s maintained by Microsoft teams and distills field practice. A great starting point for CoE teams building an enterprise quality gate (test → publish).