solidcraft-ground-rules
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external contributors in PRs, which could contain instructions embedded in project assets or documentation. Ingestion points: Pull request content including dependencies, icons, images, and product wording. Boundary markers: No explicit delimiters or instructions to ignore embedded content are provided. Capability inventory: Execution of cargo xtask commands (licences, assets, icons) and interaction with external program APIs (FreeCAD). Sanitization: No sanitization or validation of the input data is specified.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to run specific project-local CLI tools for validation and computation. Evidence: Mentions running cargo xtask and the tools/freecad-oracle/ scripting API.
Audit Metadata