performance-optimizer
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of instructional documentation and code examples for performance optimization. No malicious patterns, obfuscation, or unauthorized data access were found in the skill content.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze user-provided code, which creates a standard ingestion surface for development-oriented tasks.
- Ingestion points: User-provided source code, database schemas, and API definitions analyzed in SKILL.md.
- Boundary markers: None present in the skill instructions to separate analysis from instruction execution.
- Capability inventory: Mentions standard diagnostic tools such as node profiling and database explain plans in SKILL.md.
- Sanitization: No custom sanitization is implemented, relying on the host environment's tool safety.
Audit Metadata