skills/buildgreatproducts/product-os-public/studio-distribute-growth-experiments/Gen Agent Trust Hub
studio-distribute-growth-experiments
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core workflow of researching external, attacker-controlled content to generate marketing experiments.
- Ingestion points: The agent is instructed to use web search and social listening tools (e.g., Syncly, web search) to "Find live tactics" and "Tear down competitor distribution" (SKILL.md, Workflow Step 2). This content originates from external websites and social platforms.
- Boundary markers: The instructions lack specific delimiting markers or system-level warnings to ignore potentially malicious instructions embedded within the researched competitor data.
- Capability inventory: The skill has file-write capabilities, creating or modifying
productos/distribute/2-Growth-Experiments.mdand3-Growth-Experiments-Tracker.mdbased on its findings (SKILL.md, Workflow Step 5). - Sanitization: There are no explicit instructions to sanitize or escape the content fetched from the web before using it to generate experiment cards.
Audit Metadata