The Agent Skills Directory

[Indirect Prompt Injection Surface]: The skill is designed to process external diagnostic logs, which constitute an ingestion of untrusted data. This is appropriately mitigated by explicit instructions to the agent to treat this data as untrusted and by the use of a specialized analyzer script that only extracts allowlisted fields, preventing malicious instructions embedded in logs from being executed.- [Local Command Execution]: The skill provides Node.js, Bash, and PowerShell scripts for analyzing log files locally. These scripts are restricted to local file operations and do not perform network requests or package installations, representing a low-risk diagnostic capability.- [Azure Authentication Best Practices]: The skill includes detailed guidance on Azure authentication, correctly advising developers to use managed identities in production and providing patterns to avoid hardcoding credentials. This promotes secure development practices across various programming languages.

azure-ai