crm-icp-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from HubSpot CRM objects (deals, companies, and contacts) which could potentially contain malicious instructions intended to influence the agent's behavior during analysis.
- Ingestion points: CRM data is fetched via the
hubspot__search_crm_objectsandhubspot__get_crm_objectstools described in steps 2, 3, and 4 of SKILL.md. - Boundary markers: Absent; the instructions do not include specific delimiters or warnings to the agent to treat the contents of CRM fields as data rather than instructions.
- Capability inventory: The skill utilizes tools for HubSpot CRM access (
mcp__cogny__hubspot__*), a reporting tool (mcp__cogny__create_finding), and standard system tools (Bash,Read,Write). - Sanitization: Absent; there are no instructions to validate, escape, or filter external content before processing it in the analysis steps.
Audit Metadata