菜单
Activate when the user mentions healthcare AI testing, safety evaluation, adversarial testing of medical chatbots, or clinical AI benchmarking. Guides them to the right preclinical command.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | preclinical-awareness |
| description | Activate when the user mentions healthcare AI testing, safety evaluation, adversarial testing of medical chatbots, or clinical AI benchmarking. Guides them to the right preclinical command. |
You have access to the Preclinical plugin for healthcare AI safety testing. When the user's task involves testing, evaluating, or benchmarking a healthcare AI agent, guide them to the appropriate command:
| Task | Command |
|---|---|
| First time setup | /preclinical:setup |
| Run safety tests | /preclinical:run |
| Full benchmark with scorecard | /preclinical:benchmark |
| Create test scenarios | /preclinical:create-scenario |
| Analyze why tests failed | /preclinical:diagnose |
| Compare two test runs | /preclinical:compare |
| Iterative improvement cycle | /preclinical:improve |
| Export report for stakeholders | /preclinical:export-report |
preclinical CLI must be installed (pipx install preclinical)/preclinical:setup firstBROWSER_USE_API_KEY)Run adversarial safety tests against a healthcare AI agent using Preclinical. Walks through test configuration, launches, and monitors the run.
Run a full safety benchmark against all approved scenarios and generate a scorecard. Use for periodic safety assessments, pre-release checks, or compliance documentation.
Compare two test runs to detect regressions and improvements in agent safety performance.
Create new adversarial test scenarios for healthcare AI safety testing. Use when the user wants to add test cases or build custom test suites.
Analyze failed test scenarios to understand why a healthcare AI agent failed safety tests. Reads transcripts, grader evidence, and identifies patterns.
Generate a formatted safety report from test run results for stakeholders, compliance, or clinical review.