post-to-campaign
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has a potential attack surface for indirect prompt injection as it ingests and processes content from external LinkedIn posts.
- Ingestion points: External LinkedIn post content is retrieved using the
get_linkedin_posttool (SKILL.md, Step 2). - Boundary markers: Absent. The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the processed post text.
- Capability inventory: The skill possesses significant capabilities through the La Growth Machine MCP, including scraping LinkedIn profiles into audiences (
create_audience_from_linkedin_url) and modifying campaign messages (add_campaign_step_message,edit_campaign_message). - Sanitization: Absent. The skill directs the agent to derive the outreach 'angle' directly from the post's content, creating a path for malicious instructions in a post to be echoed into outreach sequences.
- [SAFE]: The skill incorporates safety best practices by leaving campaigns in a 'draft' state, ensuring that a human must review the content and manually launch the campaign.
Audit Metadata