skills/geminilight/mindos/find-skills/Gen Agent Trust Hub

find-skills

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses the npx skills CLI to perform searches and manage the installation of agent extensions.
  • [EXTERNAL_DOWNLOADS]: Facilitates the installation of code from third-party repositories, such as GitHub, using the npx skills add command.
  • [EXTERNAL_DOWNLOADS]: References official skill sets from well-known and established organizations like Vercel Labs and Anthropic.
  • [PROMPT_INJECTION]: Presents an indirect prompt injection attack surface where metadata or descriptions from untrusted third-party skills could be ingested by the agent during search operations.
  • Ingestion points: External skill metadata retrieved via the npx skills find command (SKILL.md).
  • Boundary markers: Not present; the agent processes third-party search results directly for presentation to the user.
  • Capability inventory: Execution of shell commands via the npx skills CLI suite.
  • Sanitization: Includes procedural checks for the agent to verify the reputation of skills (e.g., install counts and repository stars) before recommending or performing installations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 04:58 PM
Security Audit — agent-trust-hub — find-skills