skills/agentscope-ai/skills/guidance/Gen Agent Trust Hub

guidance

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes common shell commands such as find, which, awk, sed, and cat to navigate the file system, determine the project root, and retrieve the contents of markdown documentation files.
  • [EXTERNAL_DOWNLOADS]: The skill directs the agent to refer to the official vendor website http://qwenpaw.agentscope.io/ if local information is unavailable. This is a legitimate resource belonging to the skill's authoring organization.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its core function of reading and interpreting local documentation files.
  • Ingestion points: Documentation files matching *.md patterns located in directories like ~/.qwenpaw/memory/docs or website/public/docs/ (SKILL.md).
  • Boundary markers: Absent; the instructions do not specify the use of delimiters or specific commands to disregard instructions that might be contained within the documentation content.
  • Capability inventory: The skill employs shell command execution for locating and reading files across the file system (SKILL.md).
  • Sanitization: Absent; the skill reads and organizes information from the documents without explicit sanitization or filtering of the text content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:12 AM