ad-creative
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill includes functionality to analyze performance data provided via CSV, text pastes, or API outputs (as described in SKILL.md). This creates an indirect prompt injection surface where malicious instructions embedded within the performance data could influence the agent's behavior.
- Ingestion points: Performance data provided by the user or fetched through ad platform APIs (SKILL.md).
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined for the data analysis loop.
- Capability inventory: The skill possesses the ability to execute shell commands via local CLI tools and generate content for various platforms (SKILL.md).
- Sanitization: There is no mention of sanitization or validation logic for the ingested data.
- [COMMAND_EXECUTION]: The skill provides explicit instructions and examples for using local command-line tools to interact with advertising platforms (e.g.,
google-ads,meta-ads). - Evidence: The 'Tool Integrations' section in SKILL.md documents commands such as
node tools/clis/google-ads.js reports get --type ad_performance. These tools are used to fetch the data that the skill subsequently analyzes.
Audit Metadata