post-to-campaign
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes untrusted data from external LinkedIn posts to generate outreach sequences.
- Ingestion points: The skill uses the
get_linkedin_posttool (defined in the La Growth Machine MCP) to ingest the text content of a LinkedIn post provided via URL (SKILL.md, Step 2). - Boundary markers: The instructions do not specify the use of delimiters (like XML tags or clear separators) or specific "ignore embedded instructions" warnings when providing the untrusted post content to the model for analysis.
- Capability inventory: The skill possesses significant capabilities through the LGM MCP, including creating audiences (
create_audience_from_linkedin_url), duplicating campaigns (duplicate_campaign), and modifying campaign messages (add_campaign_step_message,edit_campaign_message). - Sanitization: While the skill includes copywriting rules to avoid "AI-like" patterns (references/copywriting-rules.md), it lacks specific sanitization or filtering logic to detect or neutralize malicious instructions that might be embedded within the LinkedIn post text.
Audit Metadata