web-performance-optimization
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves ingesting and acting upon data from external sources, specifically performance audit outputs (Lighthouse) and project source files (JavaScript, HTML, CSS). This creates an attack surface where an adversary could embed malicious instructions within a project's codebase or within metadata that appears in performance reports, potentially influencing the agent's subsequent optimization actions.
- Ingestion points: The skill instructs the agent to "Run Lighthouse audits", "Check bundle sizes", and "Analyze performance problems" in
SKILL.mdunder the 'How It Works' section. - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat tool outputs or analyzed code as untrusted data.
- Capability inventory: The skill empowers the agent to modify application code, install/uninstall packages via
npm, and configure build tools like Webpack. - Sanitization: No sanitization or validation logic is defined for the external inputs derived from performance tools or source code analysis.
Audit Metadata