apollo-migration-deep-dive
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The SKILL.md file executes a shell command (!npm list 2>/dev/null | head -10) at load time. This is used for environment introspection to display project dependencies but represents an automated execution point.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external CRM exports (CSV/JSON) which could contain malicious instructions designed to influence the agent's behavior during migration. 1. Ingestion points: Data is read from external files in the assess function and processed in the migration loops within SKILL.md and references/implementation-guide.md. 2. Boundary markers: The skill does not implement delimiters or provide explicit instructions for the agent to ignore potentially malicious content within the CRM records. 3. Capability inventory: The skill has access to network tools (axios for Apollo.io API requests) and file system operations (fs.writeFileSync). 4. Sanitization: The processing is focused on structural mapping and basic formatting (e.g., URL validation) rather than security-focused sanitization to neutralize embedded instructions.
Audit Metadata