azure-ai-projects-ts
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to install official Microsoft Azure SDK packages (
@azure/ai-projects,@azure/identity,@azure/monitor-opentelemetry) and the well-known@opentelemetry/apilibrary from the public NPM registry. - [DATA_EXPOSURE]: Documentation includes the
getWithCredentialsmethod for accessing connection secrets. The skill correctly identifies this as a sensitive operation that is not intended for general discovery and warns about its use. - [DYNAMIC_EXECUTION]: The skill includes configuration examples for the Azure AI code interpreter and Model Context Protocol (MCP) tools. These are documented alongside explicit warnings that they can execute code or disclose private data, directing users to verify boundaries before deployment.
- [INDIRECT_PROMPT_INJECTION]: The skill acknowledges an attack surface related to processing datasets and agent conversation inputs. It includes guidance to confirm project scope and budget before executing agent runs or dataset uploads.
Audit Metadata