apollo
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
@useclawlink/clitool via shell commands to login, connect services, and run CRM actions. This is the primary method of operation for the integration. - [EXTERNAL_DOWNLOADS]: The skill downloads and runs the
@useclawlink/clipackage from the NPM registry usingnpx. This is a vendor-supplied package required for the skill's functionality. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from the Apollo CRM, such as contact and organization details, which could contain instructions from external parties. This creates an attack surface for indirect prompt injection.
- Ingestion points: Actions like
apollo_search_contacts,apollo_people_enrichment, andapollo_organization_searchinSKILL.mdretrieve data from external sources. - Boundary markers: None implemented; however, the skill instructs the agent to read data and request user confirmation before performing write actions.
- Capability inventory: The skill can execute shell commands and perform write operations (e.g., creating deals, enrolling contacts in sequences) within the Apollo environment.
- Sanitization: No specific sanitization or validation of the retrieved CRM data is described.
Audit Metadata