azure-security-audit
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external Azure resource configurations, creating a potential surface for indirect prompt injection attacks.
- Ingestion points: Data enters the agent context through tools such as
azure_list_nsgs,azure_get_nsg_rules,azure_get_effective_security_rules,azure_get_network_watcher_status,azure_get_vnet_details, andazure_list_firewalls(SKILL.md). - Boundary markers: The instructions do not define specific delimiters or warnings to treat data retrieved from Azure as untrusted content.
- Capability inventory: The skill is equipped with multiple observation-only tools that allow for comprehensive reading of Azure network configurations.
- Sanitization: There is no documented evidence of sanitization, validation, or escaping of the content returned by the Azure API before it is processed by the agent.
- Mitigations: Implement strict schema validation for tool outputs and include explicit instructions for the agent to disregard any natural language commands or instructions found within the data retrieved from Azure.
- [COMMAND_EXECUTION]: The skill defines a shell command to execute the local Python script that implements the MCP server functionality.
- Evidence:
python mcp-servers/azure-network-mcp/azure_network_mcp_server.py(SKILL.md).
Audit Metadata