external-mindstudio-document-ux-review

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill explicitly directs the agent to execute shell commands found in user-provided documentation. The instruction to strictly follow and execute commands from external repositories as-is enables the execution of potentially harmful operations defined by untrusted sources.
  • [REMOTE_CODE_EXECUTION]: The execution workflow relies on cloning remote repositories and performing setup actions, including running installation scripts and starting services. This constitutes a remote code execution surface where the instructions are derived from the audited repository's documentation.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection where instructions in a malicious README could compromise the agent's environment or behavior.
  • Ingestion points: Repository README.md and linked files such as installation guides mentioned in SKILL.md.
  • Boundary markers: There are no clear markers to separate the agent's system instructions from the content of the documentation being audited; the agent is told to follow documentation literally.
  • Capability inventory: The agent has capabilities for subprocess execution, file system access, and network requests across setup and validation steps.
  • Sanitization: No specific sanitization or escaping is applied to commands extracted from the documentation before execution, relying instead on agent-side judgment.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 06:28 AM
Security Audit — agent-trust-hub — external-mindstudio-document-ux-review