supabase

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches configuration, changelogs, and guidelines from official Supabase domains and GitHub repositories, including https://supabase.com/changelog.md and https://mcp.supabase.com/mcp. These references target a trusted organization and are documented neutrally as intended functionality for staying updated with the platform.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the Supabase CLI (supabase db pull, supabase migration new) and MCP tools (execute_sql, get_advisors) for managing database schemas and iterating on migrations. These operations are within the scope of a database management skill and are accompanied by instructions to verify commands via --help.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides mechanisms to ingest external data by fetching documentation pages as markdown and using a documentation search tool. While this presents an indirect prompt injection surface, the risk is mitigated by directing the agent exclusively to official Supabase documentation sources.
  • Ingestion points: Documentation fetching via search_docs and markdown URL retrieval described in SKILL.md.
  • Boundary markers: Not explicitly defined for external content ingestion.
  • Capability inventory: The skill has access to SQL execution tools (execute_sql) and file system operations via the CLI.
  • Sanitization: No specific content sanitization or filtering logic is provided for the fetched documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 07:23 AM
Security Audit — agent-trust-hub — supabase