aci-change-deploy
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external systems (Cisco APIC and ServiceNow) to automate deployment logic and safety decisions, such as identifying fault deltas and triggering rollbacks.
- Ingestion points: Tool outputs from ACI
faults,health, and ServiceNowget_change_request_details(SKILL.md). - Boundary markers: No explicit delimiters or instructional warnings for the processing of tool-returned data were observed.
- Capability inventory: The skill can execute policy changes on Cisco APIC and manage task states in ServiceNow.
- Sanitization: No explicit validation or sanitization logic for the structured tool output is documented in the instructions.
- [COMMAND_EXECUTION]: The skill guides the agent to construct and execute shell commands that include environment variables for both script paths and fabric credentials (e.g.,
$ACI_USERNAME,$ACI_PASSWORD). - [DYNAMIC_EXECUTION]: The skill uses a dynamic execution pattern where tool script paths are retrieved from environment variables (
$ACI_MCP_SCRIPT,$SERVICENOW_MCP_SCRIPT,$GAIT_MCP_SCRIPT) and assembled into command strings at runtime.
Audit Metadata