programming-advisor
Pass
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill is designed to read project manifest files (e.g., package.json, requirements.txt, Cargo.toml) and directory structures to understand the user's framework and dependency context. This is a legitimate functional requirement for providing relevant integration advice.
- [COMMAND_EXECUTION]: The skill generates shell commands for various package managers (npm, pip, cargo, etc.) to assist users in installing recommended libraries. These commands are presented to the user as instructions and are not executed automatically by the agent.
- [EXTERNAL_DOWNLOADS]: The skill utilizes web search to identify third-party solutions. It references well-known technology companies and official package registries (e.g., Stripe, Auth0, Clerk, Vercel) which are established and trusted sources for software development.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted data from web search results and project files. This is documented as an inherent architectural property of agentic skills that ingest external data for analysis.
Audit Metadata