recoup-internal-consulting-content-idea-generator
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external LinkedIn interactions, which creates a potential vector for indirect prompt injection if user-generated content is processed.
- Ingestion points: The skill reads performance signals and engagement data from the
integrations/linkedin/engagement/directory. - Boundary markers: The instructions do not provide specific delimiters or instructions for the agent to disregard potential commands embedded within the engagement data.
- Capability inventory: The skill is authorized to write variant logs and signals to the
signals/directory. - Sanitization: No sanitization, filtering, or validation processes are defined for the engagement data before it is analyzed by the agent.
Audit Metadata