NYC
skills/vm0-ai/vm0-skills/supabase/Gen Agent Trust Hub

supabase

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • COMMAND_EXECUTION (SAFE): The skill uses bash -c to execute curl commands. This is used to work around environment variable handling issues and is the core intended functionality of the skill.
  • DATA_EXFILTRATION (SAFE): All network operations are directed to the user-configured ${SUPABASE_URL}. No evidence of unauthorized data transmission or exfiltration to third-party domains was found.
  • CREDENTIALS_UNSAFE (SAFE): The skill correctly identifies sensitive keys using the vm0_secrets and vm0_vars metadata fields. No hardcoded API keys or credentials are present in the documentation or command examples.
  • INDIRECT_PROMPT_INJECTION (LOW): While the skill reads request bodies from /tmp/supabase_request.json, it does not automatically process or execute instructions contained within that data. The primary risk is the agent being tricked into performing unintended database operations, which is inherent to any database-interaction skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:08 PM