shunk031-herdr-tab-status
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the herdr command-line utility and jq for environment state management and status reporting. It employs a verify-after-write logic to ensure labels are updated correctly and uses jq argument binding to pass parameters safely.
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates external information like task descriptions and pull request identifiers into tab labels and agent messages.
- Ingestion points: Runtime data enters the context via placeholders like and in SKILL.md.
- Boundary markers: No explicit delimiters are used for the interpolated strings in the tab labels.
- Capability inventory: Capabilities are limited to herdr tool invocations and jq processing.
- Sanitization: Shell snippets demonstrate proper use of double quotes and jq --arg for variable interpolation, mitigating command injection risks.
Audit Metadata