supabase-advanced-troubleshooting

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use psql and supabase CLI for advanced database management and project configuration. This includes high-privilege operations such as terminating database backend processes (pg_terminate_backend), creating targeted indexes, and modifying database extensions. These tools are scoped within the allowed-tools configuration to limit the execution environment.
  • [DATA_EXFILTRATION]: The skill provides procedures and scripts (such as advanced-supabase-debug.sh and buildDiagnosticBundle) for aggregating diagnostic information, including database statistics, application logs, and system metrics. While intended for manual troubleshooting and support escalation, this involves the consolidation of sensitive operational data. The code correctly utilizes environment variables for authentication and does not contain hardcoded secrets.
  • [PROMPT_INJECTION]: The skill provides instructions for the agent to analyze the content of pg_stat_statements, which records raw SQL queries executed in the database. This creates a standard surface for indirect prompt injection if malicious instructions are embedded within the SQL queries being analyzed. The risk is inherent to the diagnostic purpose and is mitigated by the agent's general safety guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 01:20 AM
Security Audit — agent-trust-hub — supabase-advanced-troubleshooting