agent-implementer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [NO_CODE] (SAFE): The skill consists entirely of Markdown documentation and TypeScript code snippets intended for reference and manual implementation. No scripts or executables are provided or invoked.
- [PROMPT_INJECTION] (SAFE): No instructions designed to override agent behavior, bypass safety filters, or extract system prompts were detected. The content focuses on architectural guidance for creating agents.
- [DATA_EXFILTRATION] (SAFE): There are no network requests (curl, wget, fetch) or access to sensitive local file paths. No hardcoded credentials or API keys are present.
- [REMOTE_CODE_EXECUTION] (SAFE): No external dependencies, package installations, or remote script downloads are included.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill defines prompt templates that interpolate untrusted data (e.g.,
{{task}}and{{context}}). While this creates a potential attack surface for the agents built using these templates, the skill itself does not process this data or execute the resulting prompts. Developers following these templates should implement their own sanitization and boundary markers.
Audit Metadata