write-korean-business

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the uv run utility to process documents. This pattern poses a risk if user-controlled file paths are passed as arguments to the scripts without proper validation or escaping.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it ingests and reconstructs external materials (referenced in SKILL.md). Evidence Chain:\n
  • Ingestion points: Source material for reconstruction enters the context when user requests to "reconstruct from material" as per SKILL.md.\n
  • Boundary markers: Absent; there are no instructions to use delimiters or ignore instructions within the source data.\n
  • Capability inventory: The skill uses shell execution capabilities via uv run in SKILL.md.\n
  • Sanitization: Absent; there is no mention of escaping or filtering external content.\n- [DYNAMIC_EXECUTION]: The instructions in SKILL.md describe executing external Python scripts from a relative directory path ($CORE_SKILL_DIR/scripts/). This constitutes dynamic execution of code that is not bundled with the skill itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 11:17 PM
Security Audit — agent-trust-hub — write-korean-business