pipelines-build-summary
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
curlto interact with the official Azure DevOps REST API atdev.azure.com. The instructions include specific guidance on usingcurl.exefor Windows compatibility and require URL encoding for user-provided project and pipeline names to mitigate shell injection risks. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources, specifically build logs and commit messages retrieved from Azure DevOps.
- Ingestion points: API responses containing timeline records, build logs, and commit metadata as described in
SKILL.md. - Boundary markers: The skill directs the agent to present findings in structured tables and code blocks to maintain separation between data and instructions.
- Capability inventory: Execution capabilities are limited to
curlfor making authenticated API requests to known Microsoft domains. - Sanitization: The skill mandates URL encoding for variables interpolated into the API query strings and paths.
Audit Metadata