pick-next-issue
Pass
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
ghCLI tool to list and view GitHub issues. These commands are used to fetch public or repository-specific metadata and descriptions. - [SAFE]: No malicious behaviors such as data exfiltration, credential harvesting, or unauthorized persistence were identified. The tool's operations are confined to read-only interactions with GitHub issues.
- [SAFE]: Indirect Prompt Injection Surface:
- Ingestion points: External content enters the agent's context through
gh issue list(Step 1) andgh issue view(Step 4). - Boundary markers: None identified; fetched content is passed directly to the planning step.
- Capability inventory: The skill executes shell commands (
gh) and invokes another skill (/turboplan). - Sanitization: None mentioned for the issue content being processed. While this presents a surface for indirect prompt injection, it is standard for tools interacting with issue trackers and does not elevate the risk level for this specific implementation.
Audit Metadata