supabase

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external documentation from Supabase and GitHub, which presents a potential surface for indirect instructions to influence agent actions.\n
  • Ingestion points: Documentation fetching via search_docs and markdown conversion of Supabase URLs as described in SKILL.md.\n
  • Capability inventory: The agent is granted capabilities to execute SQL via execute_sql and manage the file system using the Supabase CLI.\n
  • Boundary markers: Instructions do not explicitly require the use of delimiters or ignore-instructions for retrieved documentation content.\n
  • Sanitization: There is no mention of filtering or sanitizing documentation content before it is processed by the agent.\n- [COMMAND_EXECUTION]: The skill provides detailed instructions for executing Supabase CLI commands (supabase) to manage database schema changes, migrations, and project health checks.\n- [EXTERNAL_DOWNLOADS]: The skill fetches documentation and submits feedback to official Supabase infrastructure, including supabase.com, mcp.supabase.com, and the supabase/agent-skills repository on GitHub. These are recognized as trusted vendor resources.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:53 AM
Security Audit — agent-trust-hub — supabase