skills/lijigang/ljg-skills/ljg-is/Gen Agent Trust Hub

ljg-is

Pass

Audited by Gen Agent Trust Hub on Oct 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-defined concepts and optional raw materials (such as external documents or URLs) to create explanatory notes. It lacks explicit boundary markers or sanitization instructions for these inputs during prompt interpolation, which creates a surface for indirect prompt injection where malicious content in the processed materials could attempt to influence the agent's behavior.
  • [COMMAND_EXECUTION]: As part of its standard workflow defined in Workflows/TraceCreation.md, the skill executes a local script (Tools/ValidateNote.ts) using the bun runtime to verify the structure and quality of generated notes. While this is an internal tool provided with the skill, it involves the execution of shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 30, 2026, 05:09 PM