linkedin-ads
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: Authentication is managed by the Membrane platform, avoiding the exposure of sensitive API keys or tokens in the skill code or agent environment.
- [COMMAND_EXECUTION]: The skill uses the
membraneCLI tool to list, create, and update LinkedIn Ads entities like campaigns and accounts. - [EXTERNAL_DOWNLOADS]: Installs the
@membranehq/clipackage, which is a legitimate vendor tool for interacting with the Membrane ecosystem. - [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection from LinkedIn Ads data. 1. Ingestion points: Data returned from the LinkedIn Ads API (e.g., campaign names). 2. Boundary markers: Absent. 3. Capability inventory: Command execution via
membraneCLI including write/delete operations. 4. Sanitization: Not specified in instructions.
Audit Metadata