The Agent Skills Directory

[SAFE]: The skill provides comprehensive and legitimate architectural guidance for enterprise Supabase deployments, including RLS policy templates and migration strategies.
[EXTERNAL_DOWNLOADS]: The documentation examples reference standard and well-known libraries from trusted CDNs such as deno.land and esm.sh (e.g., @supabase/supabase-js).
[COMMAND_EXECUTION]: The skill is authorized to use Bash and Write tools, which are powerful capabilities. While these are appropriate for its purpose as a development assistant, they represent an exploitable surface if the agent is misled.
[PROMPT_INJECTION]: The skill exhibits an Indirect Prompt Injection surface (Category 8).
Ingestion points: Processes ApplicationRequirements from users and fetches external documentation via the mcp__context7__get-library-docs tool.
Boundary markers: None explicitly defined in the logic to isolate external data from internal instructions.
Capability inventory: Includes Bash, Write, Edit, and WebFetch, allowing for file modification and code execution.
Sanitization: No specific sanitization or validation logic is present in the skill's instructions for handling external data.

moai-baas-supabase-ext