design-principles
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external codebase files, which serves as an ingestion point for untrusted data.
- Ingestion points: Project source files, README.md, and CONTEXT.md (SKILL.md Step 1).
- Boundary markers: Absent; no specific instructions are provided to ignore embedded commands in the audited code.
- Capability inventory: Executes shell commands (git, ls, markdownlint) and performs file system operations (SKILL.md Step 5).
- Sanitization: Absent; the code is read and processed directly by the agent's reasoning.
- [COMMAND_EXECUTION]: The skill executes local shell commands for environmental checks and audit tasks.
- Evidence: Commands include
git --version,lsfor build file detection, andmarkdownlintfor report validation (SKILL.md).
Audit Metadata