simplified-technical-english
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-supplied technical prose (procedures and descriptions) which introduces a standard injection surface.\n
- Ingestion points: User-provided text is processed by instructions in SKILL.md, references/procedures-and-safety.md, and analyzed by scripts/analyze_ste_surface.py.\n
- Boundary markers: The skill employs explicit preservation rules for protected literals and recommends the use of start/end markers (e.g., ) to delimit processed text.\n
- Capability inventory: The included Python script analyze_ste_surface.py performs regex-based linguistic analysis and outputs results to stdout. It does not perform network operations, file system writes, or dynamic code execution.\n
- Sanitization: The diagnostic script masks fenced code, inline code, and quoted text using regular expressions to prevent them from influencing linguistic metrics, providing a layer of structural isolation.\n- [SAFE]: Analysis of the skill's instructions and supporting scripts confirms no malicious intent, obfuscation, or unauthorized data access. External references point to official documentation for the ASD-STE100 standard.
Audit Metadata