pangu
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and modify user-supplied text, which represents an indirect prompt injection surface.
- Ingestion points: Processes prose, documents, code comments, and commit messages as described in SKILL.md.
- Boundary markers: Contains explicit instructions to preserve literal content (code, quotes, logs) and structural syntax (URLs, paths, Markdown markers) byte-for-byte.
- Capability inventory: No command execution, file system access, or network capabilities are present in the skill instructions or metadata.
- Sanitization: Relies on the agent's ability to distinguish between natural language prose and technical tokens/code based on the provided rules.
- [SAFE]: The skill consists entirely of natural language instructions and formatting rules. It does not perform any high-risk operations and no malicious patterns or obfuscation techniques were detected.
Audit Metadata