prospect-research-to-cadence
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it incorporates external, untrusted data into its automated workflow.
- Ingestion points: The skill reads external data from
apollo_organizations_job_postingsand general web search results to inform prospect research and personalization (Stage 1 inSKILL.md). - Boundary markers: The instructions do not define specific delimiters or instructions to prevent the agent from following commands that might be embedded in the external data during the drafting process.
- Capability inventory: The skill has access to sensitive actions such as creating CRM contacts (
apollo_contacts_create), enrolling users in automated email sequences (apollo_emailer_campaigns_add_contact_ids), and creating email drafts (gmail_create_draft). - Sanitization: The workflow includes a mandatory 'Stage 3: Approve' step where the user reviews and approves all generated drafts before the skill proceeds with contact creation or sequence loading, which serves as a primary security control.
Audit Metadata