programming-principles
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted data such as source code and GitHub metadata, which creates a potential surface for indirect prompt injection.
- Ingestion points: The 'Principles-Based Code Assessment Workflow' (
references/code-assessment-workflow.md) instructs the agent to map repository trees, read every source file, and fetch external data via the GitHub CLI (gh pr view,gh issue list). - Boundary markers: The skill does not provide specific instructions or delimiters to help the agent distinguish between its own internal instructions and potentially malicious instructions embedded within the code or issues being analyzed.
- Capability inventory: The workflow suggests the use of file system tools (
find), GitHub integration (gh), and repository extraction tools. - Sanitization: The skill lacks instructions for sanitizing or escaping the content retrieved from analyzed repositories before it is processed by the agent.
- [SAFE]: The skill consists entirely of instructional Markdown documentation and distillation of programming principles. It contains no executable scripts, hardcoded credentials, remote dependencies, or obfuscated content. All external references point to reputable sources such as official GitHub repositories.
Audit Metadata