paid-ads
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest data from an external file (
.claude/product-marketing-context.md) and user-provided inputs (landing page URLs, campaign goals) without explicit boundary markers or sanitization. This creates a surface for indirect prompt injection if those sources are compromised. - Ingestion points: Reads project context from
.claude/product-marketing-context.md(defined in SKILL.md). - Boundary markers: None identified for external file reading or user input interpolation.
- Capability inventory: No dangerous capabilities such as subprocess execution, file writing, or network operations are defined or requested in these files.
- Sanitization: No sanitization or validation of the external content is performed.
- [NO_CODE]: The skill consists solely of Markdown documentation, templates, and reference guides. It lacks any executable scripts or binary files, which minimizes the risk of direct malicious code execution.
- [SAFE]: The skill refers to standard platform integration tools (Google Ads, LinkedIn, etc.) through internal documentation links and does not contain any hardcoded credentials, obfuscation, or unauthorized network activity.
Audit Metadata