The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data in the form of business requirements, PRDs, or specification documents to generate database schemas.
Ingestion points: The agent is instructed to read requirements documents in SKILL.md (Step 1) to extract entities and relationships.
Boundary markers: The instructions lack specific delimiters or warnings to ignore potentially malicious instructions embedded within the requirements documents.
Capability inventory: The skill possesses the Write tool to create file system artifacts and Bash capabilities via the Supabase CLI (npx supabase) to modify database states and push migrations.
Sanitization: There is no evidence of sanitization or validation of the input text before it is used to inform the generation of SQL migrations or RLS policies.
[COMMAND_EXECUTION]: The skill relies on the execution of shell commands through the Supabase CLI (npx supabase) for core functionality.
Evidence: SKILL.md uses npx supabase migration new, npx supabase db reset, npx supabase db push, and npx supabase gen types. These are legitimate development commands for a well-known service.

supabase-schema-from-requirements