Skills
Navigate
skills/leverage-computer/skills/shared-agents-md/Gen Agent Trust Hub

shared-agents-md

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes the leverage CLI tool to manage workspace instructions and sessions, which is consistent with the author's identity (leverage-computer).
  • [EXTERNAL_DOWNLOADS]: The instructions recommend installing an additional skill from the author's official GitHub repository (leverage-computer/skills). This is a legitimate way to extend the agent's capabilities within the same ecosystem.
  • [COMMAND_EXECUTION]: The skill invokes commands like leverage whoami and leverage sessions new. These are standard operational procedures for the platform being used and do not involve unauthorized privilege escalation or dangerous parameters.
  • [INDIRECT_PROMPT_INJECTION]: While the skill involves merging and processing user-provided markdown files (CLAUDE.md into AGENTS.md), this is its primary stated purpose. Standard guardrails for processing external text apply, and the skill does not introduce unique vulnerabilities in how it handles this data.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 06:41 AM