find-skills

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses the npx skills CLI to perform searches and manage package installations.\n- [EXTERNAL_DOWNLOADS]: Fetches skill definitions and packages from the skills.sh registry and various GitHub repositories, including Vercel Labs' public repository.\n- [REMOTE_CODE_EXECUTION]: The npx skills add command is used to download and install executable code. The skill instructions suggest using the -y flag to skip installation confirmation, which could facilitate the silent execution of third-party code.\n- [PROMPT_INJECTION]: The skill processes external search results from the CLI, creating a surface for indirect prompt injection.\n
  • Ingestion points: Search results from the npx skills find command (SKILL.md).\n
  • Boundary markers: Absent; there are no instructions to the agent to validate or sanitize content returned from external searches.\n
  • Capability inventory: The agent has the ability to write files and execute shell commands (npx skills add) based on the untrusted search output (SKILL.md).\n
  • Sanitization: Absent; search results are used directly for presentation or installation without filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 12:15 AM