optimize-web-animations
Pass
Audited by Gen Agent Trust Hub on Jul 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands for repository management and build verification, including
git status,git diff,git read-tree,git commit,npm run lint, andnpm run build. These operations are used according to standard development practices for patching and verifying code. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because its browser profiling scripts (
references/browser-profiling.md) extract and report text content, class names, and page titles from arbitrary web pages to the agent. This external data could potentially influence the agent's behavior if a website contains malicious instructions hidden in metadata or page text. - Ingestion points:
textContent,className, anddocument.titleare extracted from targeted web pages inreferences/browser-profiling.mdandsamplePerformancePagefunction. - Boundary markers: No specific delimiters or "ignore instructions" directives are applied to the external content before it is incorporated into the agent's context.
- Capability inventory: The skill possesses the capability to modify local files, execute shell commands, and manage git commits in
SKILL.md. - Sanitization: The skill performs basic string slicing on extracted content but does not sanitize or validate it for embedded instructions.
Audit Metadata