icp-modeller
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions prompt the agent to ingest and process potentially untrusted external data sources, creating an attack surface for indirect prompt injection. \n
- Ingestion points: CRM exports, customer lists, campaign history, and sales call transcripts provided by the user, as well as external competitor reviews and G2 data mentioned in the 'Before You Start' and 'Pre-Revenue Workflow' sections. \n
- Boundary markers: The skill does not define specific delimiters or instructions to ignore potentially malicious embedded content within the ingested data. \n
- Capability inventory: The skill is purely instructional and does not define or utilize any specific tools, scripts, or network capabilities. \n
- Sanitization: The instructions lack guidance for sanitizing, escaping, or validating the external content before it is processed by the agent. \n- [NO_CODE]: The skill consists exclusively of markdown documentation and does not include any scripts, executable binaries, or system command calls.
Audit Metadata