cro
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references an external RevOps inbound-routing playbook hosted on GitHub (
github.com/coreyhaines31/marketingskills/blob/main/skills/revops/references/inbound-routing.md) for additional form optimization guidance. As a reference link on a well-known service, it does not pose a direct execution risk but points to third-party content. - [INDIRECT_PROMPT_INJECTION]: The skill is configured to ingest untrusted data from the user's workspace to gain product marketing context, which introduces an attack surface for indirect prompt injection.
- Ingestion points: The instructions prompt the agent to read
.agents/product-marketing.md,.claude/product-marketing.md, orproduct-marketing-context.mdin the current workspace. - Boundary markers: Absent; there are no specified delimiters or 'ignore embedded instructions' warnings provided to distinguish workspace data from the skill's own core logic.
- Capability inventory: The skill is primarily instructional and analytical, focusing on marketing copy and layout; it lacks dangerous capabilities such as shell command execution, file-writing, or automated network requests.
- Sanitization: Absent; the skill does not define any validation, filtering, or escaping for the content of the marketing files it processes.
Audit Metadata