aid-impl-plan
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed entirely of markdown documentation, YAML/JSON templates, and reference guides. No executable scripts, binaries, or active code components are present in the skill files.
- [PROMPT_INJECTION]: The instructions establish 'Iron Rules' and 'Golden Rules' to define information boundaries for an AI agent. These serve as functional constraints for the implementation planning process and do not attempt to override safety protocols or extract system prompts.
- [DATA_EXFILTRATION]: No hardcoded credentials, sensitive file access, or suspicious network operations were detected. Mentions of Jira API calls are illustrative documentation examples using placeholder project keys (e.g., 'AID').
- [REMOTE_CODE_EXECUTION]: There are no commands for external package installation or remote script execution. The skill operates purely as a data processing and formatting tool for local documentation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external documents such as PRDs and Tech Specs to generate planning artifacts. While this constitutes an attack surface, the skill implements 'Sanitization Rules' in
references/criteria-generator.mdand 'Verification Agents' inreferences/phase3-methodology.mdto mitigate the risk of malicious content from source documents influencing the final output.
Audit Metadata