boards-backlog-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 perform legitimate REST API calls to Azure DevOps for project, team, and work item data retrieval. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from an external source (Azure DevOps work item titles, states, and paths) which could potentially contain malicious formatting or instructions.
- Ingestion points: Data is ingested from Azure DevOps REST API responses in steps 2, 3, 5, and 6.
- Boundary markers: The skill uses Markdown tables and header structures to separate data fields, providing some structural delimiters.
- Capability inventory: The skill has network access via
curland file-writing capabilities (for JSON payloads in PowerShell environments). - Sanitization: No explicit sanitization or escaping of work item content is mentioned before interpolation into the final output.
Audit Metadata