context-building
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill aggregates data from potentially untrusted sources, including user input and external campaign results (CSV exports), to create a central context file consumed by other GTM agents.
- Ingestion points: Data is ingested from user-provided campaign scope and external campaign results files (e.g., Instantly exports) as described in SKILL.md.
- Boundary markers: The schema for the generated context file does not include explicit delimiters or instructions to ignore embedded commands within the dynamic fields.
- Capability inventory: The skill performs file-write operations to the local filesystem path campaigns/{campaign-slug}/context.md.
- Sanitization: The instructions do not define any sanitization, escaping, or validation steps for the content pulled from external results or user input before it is written to the context file.
Audit Metadata