# Microsoft — Introducing run-assert-eval: Find the risk, fix it, prove it

- Company: Microsoft (microsoft.com)
- Announced: 2026-09-24T17:28:30+00:00
- Category: developer-tool-release
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://commandline.microsoft.com/run-assert-eval-responsible-ai-agent-risk-discovery-at-runtime/
- Record: https://forck.live/items/19126-introducing-run-assert-eval-find-the-risk-fix-it-prove-it
- Subject: Foundry / Responsible AI

Microsoft introduced run-assert-eval, a VS Code skill that discovers AI agent risks, measures failure rates, generates runtime policies from findings, and reruns evaluations to verify fixes, all from a single prompt. The skill integrates Clarity, ASSERT, and Agent Control Specification to automate governance and preserve evaluation integrity.

## Evidence

Verbatim from https://commandline.microsoft.com/run-assert-eval-responsible-ai-agent-risk-discovery-at-runtime/:

> From a single prompt in VS Code, it discovers the risks that matter for a given agent, measures how often the agent fails, generates runtime policy directly from those findings, and reruns the same evaluation to prove whether the fix worked.

---

Record: https://forck.live/items/19126-introducing-run-assert-eval-find-the-risk-fix-it-prove-it
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md