renewal-risk-radar
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPERSISTENCEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
claycommand-line interface to verify user identity, list platform routines, and manage workflow nodes. - [DATA_EXFILTRATION]: Accesses sensitive account and contact data from connected CRM systems. The results are formatted into a digest and sent to external destinations like Slack, email, or internal platform tables as configured by the user.
- [PERSISTENCE]: Configures a persistent, recurring scheduled trigger (defaulting to Monday at 07:00) to execute the risk analysis workflow unattended.
- [DYNAMIC_EXECUTION]: Employs "code nodes" to perform data manipulation, signal merging, and ranking logic. This involves executing custom logic within the platform's execution environment.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external CRM objects and public enrichment signals (Step 1, Step 3), creating a potential attack surface for indirect prompt injection if those external sources contain malicious instructions. The instructions lack explicit mention of data sanitization or boundary markers for these inputs. The skill's capabilities include enrichment calls, network communications to Slack/email, and execution within code nodes.
Audit Metadata