write-korean-public

Warn

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill instructs the agent to execute scripts from an external sibling directory (natural-korean/scripts/) that is not part of the skill package, representing dynamic loading from a computed path.
  • [COMMAND_EXECUTION]: The skill uses 'uv run' to execute shell commands, interpolating file path variables directly into the command line which could lead to arbitrary command execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided file paths and content in a shell context. 1. Ingestion points: User-provided document paths in SKILL.md. 2. Boundary markers: None provided for command interpolation. 3. Capability inventory: Shell command execution via 'uv run' in SKILL.md. 4. Sanitization: None; the skill does not provide instructions to sanitize or validate the file path variable before shell interpolation.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 27, 2026, 11:17 PM
Security Audit — agent-trust-hub — write-korean-public