ljg-structure

Pass

Audited by Gen Agent Trust Hub on Oct 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the execution of shell commands (date) to generate timestamps for file naming and organization.
  • [COMMAND_EXECUTION]: The skill instructs the agent to write generated content to the local filesystem at a specific path (~/Context/).
  • [INDIRECT_PROMPT_INJECTION]: Ingestion points: The workflow ingests untrusted information from the user to perform structural analysis in Workflows/FindStructure.md. Boundary markers: No specific boundary markers or instructions are provided to the agent to ignore potentially malicious instructions embedded in the user input. Capability inventory: The agent has permissions to execute shell commands and write to the local filesystem across its operational scripts. Sanitization: The subject variable ({主题}), sourced directly from user input, is interpolated into a file path in SKILL.md without validation or sanitization, creating a potential surface for path manipulation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 27, 2026, 02:35 AM