big-idea
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by instructing the agent to search the web for product facts and user-provided audience truths.
- Ingestion points: Web search results (Workflow Step 2) and user briefs are incorporated into the creative process.
- Boundary markers: The skill uses Markdown headers to structure output but does not include explicit delimiters to prevent the agent from obeying instructions potentially hidden in search results.
- Capability inventory: The skill is restricted to text generation and does not call for system-level operations like file writing or shell command execution with the ingested data.
- Sanitization: No sanitization steps are defined, but the risk is minimized by the skill's creative-focused scope.
- [EXTERNAL_DOWNLOADS]: The README provides an installation command that fetches the skill from the author's official repository ('thatjuan/agent-skills'). This is a standard and transparent distribution method for the platform.
Audit Metadata