ad-creative
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands and CLI scripts, such as
node tools/clis/google-ads.js, to retrieve performance reports and manage advertising campaigns. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted performance data (CSV, text pastes, or API outputs) for analysis.
- Ingestion points: Performance data ingested during the 'Iterate from Performance Data' workflow in
SKILL.mdandUPSTREAM_SOURCE.md. - Boundary markers: Absent; the instructions do not include specific delimiters or warnings to ignore instructions embedded within the processed data.
- Capability inventory: Shell command execution (Node.js), advertising platform API interactions (Google Ads, Meta, LinkedIn, TikTok), and file system read access for marketing context files.
- Sanitization: No data sanitization, validation, or escaping procedures are defined for the ingested performance data.
Audit Metadata