using-aidlc
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of various local PowerShell scripts (e.g., 'init-task-pack.ps1', 'test-phase-c-gate.ps1', 'validate-task-branch.ps1') to manage task lifecycles and validate process gates.
- [INDIRECT_PROMPT_INJECTION]: The routing logic is dependent on reading and interpreting content from external markdown files such as 'raw.md' and 'solution.md', which provides an entry point for potential instructions embedded in processed data.
- Ingestion points: Project files located in '{FEATURE_DIR}/requirements/' and other stage-specific directories.
- Boundary markers: The agent uses structural markers like YAML frontmatter and specific Markdown headers (e.g., '#impact-analysis') to identify routing parameters.
- Capability inventory: Access to system commands (Git) and execution of local maintenance scripts.
- Sanitization: No specific sanitization of the markdown file content is documented before the information is used to drive the workflow transitions.
Audit Metadata