draft-creator-campaign-brief
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external content from LinkedIn profiles and recent posts (Step 3) to "read their voice" and provide examples. This creates a surface for indirect prompt injection where malicious instructions embedded in a creator's post could influence the agent's behavior during brief generation.
- Ingestion points: LinkedIn profile data and recent posts fetched via analytics tools or manual input as described in Step 0 and Step 3 of SKILL.md.
- Boundary markers: Absent. The instructions do not provide delimiters or guidance for the agent to distinguish between data and potentially malicious instructions within the fetched content.
- Capability inventory: The skill utilizes tools to read from LinkedIn/analytics sources and brand records, and it writes the final output to document or slides tools (Step 4).
- Sanitization: Absent. There is no mention of filtering, escaping, or validating the content retrieved from external social media profiles before it is processed and included in the campaign brief.
Audit Metadata