multichannel-campaign-builder
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided campaign briefs to generate outreach messages. While this creates a surface for indirect prompt injection, it is mitigated by several security controls.
- Ingestion points: User input is gathered in Step 1 of
SKILL.md. - Boundary markers: The skill contains strict formatting instructions to limit output to specific deliverables.
- Capability inventory: The skill uses MCP tools (such as
edit_campaign_messageandadd_campaign_step_messageinreferences/lgm-campaign-create.md) to write content to the La Growth Machine service. - Sanitization:
references/lgm-campaign-create.mdmandates HTML escaping for all generated content and notes that the external platform validates inputs. Additionally, all campaigns are created as drafts requiring manual user approval.
Audit Metadata