paid-ads-chatgpt
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided data described as a 'verified ChatGPT Ads export'. This ingestion point creates an attack surface where malicious instructions could be embedded in the provided export to influence the agent's behavior during campaign planning or performance reviews.
- Ingestion points: The skill accepts user-provided campaign objectives, audience details, or 'verified ChatGPT Ads exports' as inputs, as specified in the description and argument-hint.
- Boundary markers: No clear delimiters or instructions to ignore embedded commands within the untrusted export data are present in this wrapper file.
- Capability inventory: While this wrapper file does not directly invoke tools, it instructs the agent to adopt a new workflow from a secondary local file which likely contains the primary logic and tool access for the skill.
- Sanitization: No logic for sanitizing, validating, or escaping the content of the external export is provided in the instructions.
Audit Metadata