Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →wasabeef/switch-to-expert-role-for-specialized-analysis
Switch to expert role for specialized analysis. Trigger with "switch role", "as a security expert", "as an architect", "as a frontend specialist".
Passed 8 of 21 test cases with the skill, and 3 of 21 without it.
and 2 more in the full report
5 fixed3 passed either way14 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
We have finished our specialized analysis session and wish to return to standard conversational assistant mode.
| Checked | Without skill | With skill |
|---|---|---|
| Includes the command `/role default`. | Missed | Passed |
The instructions your model reads when the skill loads.
Switch to a specific role to perform specialized analysis or work.
/role <role_name> [--agent|-a]
--agent or -a: Execute as a sub-agent (recommended for large-scale analysis)
security: Security audit expert (OWASP Top 10, threat modeling, Zero Trust principles, CVE matching)performance: Performance optimization expert (Core Web Vitals, RAIL model, phased optimization, ROI analysis)analyzer: Root cause analysis expert (5 Whys, systems thinking, hypothesis-driven, cognitive bias countermeasures)frontend: Frontend and UI/UX expert (WCAG 2.1, design systems, user-centered design)mobile: Mobile development expert (iOS HIG, Android Material Design, cross-platform strategy)backend: Backend and server-side expert (RESTful design, scalability, database optimization)reviewer: Code review expert (readability, maintainability, performance, refactoring proposals)architect: System architect (Evidence-First design, MECE analysis, evolutionary architecture)qa: Test engineer (test coverage, E2E/integration/unit strategy, automation proposals)# Switch to security audit mode (normal)
/role security
"Check the security vulnerabilities of this project"
# Run security audit as a sub-agent (large-scale analysis)
test cases passed, vs 3 / 21 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.