connections-optimizer
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external social media profiles which could contain malicious instructions designed to manipulate the agent's network analysis logic or message drafting.\n
- Ingestion points: Following and connection inventory and activity data retrieved from X and LinkedIn platforms.\n
- Capability inventory: Drafting DMs, LinkedIn messages, and Apple Mail emails; generating network pruning and follow recommendations.\n
- Boundary markers: The instructions lack explicit delimiters or instructions to ignore embedded commands within the ingested social media profile data.\n
- Sanitization: There is no description of filtering, sanitizing, or validating external profile content before it is processed for scoring or interpolation into outreach drafts.
Audit Metadata