skill-dev-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill manages a workflow where untrusted user input (rough ideas) and external data (GitHub issue descriptions) are used as inputs for subsequent automated steps like planning and skill implementation.
- Ingestion points: User-provided ideation input in Step 1 and issue content retrieved in Step 3.
- Boundary markers: Absent; the instructions do not define delimiters or specific warnings to ignore embedded instructions within the processed data.
- Capability inventory: The workflow performs file system writes (manifests and documentation), executes shell commands via the GitHub CLI, and performs Git operations (branching, committing, and pull requests).
- Sanitization: Absent; the orchestrator passes text content directly between specialized skills without validation or escaping logic.
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (
gh) to perform repository management tasks. - Evidence: Step 9 uses
gh sub-issue list <parent-issue-number>to retrieve data and close issues programmatically.
Audit Metadata