aci-change-deploy

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands using python3 to run local scripts defined by environment variables like $MCP_CALL, $ACI_MCP_SCRIPT, and $SERVICENOW_MCP_SCRIPT. This is the primary mechanism for interacting with ACI and ServiceNow tools.
  • [CREDENTIALS_UNSAFE]: The workflow passes credentials ($ACI_USERNAME, $ACI_PASSWORD) as environment variables within shell command strings. While these are variables and not hardcoded secrets, passing them in shell strings can expose them to process monitors or command history logs.
  • [INDIRECT_PROMPT_INJECTION]: The agent's decision-making logic, such as whether to proceed with changes or initiate a rollback, depends on data ingested from external sources like ServiceNow (Change Request status) and Cisco APIC (fault counts and health scores).
  • Ingestion points: Data returned by get_change_request_details, faults, and health tools.
  • Boundary markers: No specific delimiters or safety instructions are provided to the agent to distinguish between trusted control data and potentially malicious content within CR descriptions or fault messages.
  • Capability inventory: The skill possesses the ability to modify network policy (tenants_post, fvAEPg_post, etc.) and update ServiceNow tickets.
  • Sanitization: The skill does not explicitly describe sanitization or validation of the data retrieved from ServiceNow or APIC before using it to drive logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 12:31 AM