The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requests access to the Bash tool, providing the agent with the capability to execute shell commands. While the provided documentation uses this for legitimate database management tasks, the tool itself represents a high-privilege capability.
[EXTERNAL_DOWNLOADS]: The skill leverages WebFetch and the mcp__context7__get-library-docs tool to pull documentation and latest features from external sources. These operations target well-known documentation repositories and official library indices.
[EXTERNAL_DOWNLOADS]: Code examples for Edge Functions include external imports from deno.land and esm.sh, which are standard, well-known registries for Deno and JavaScript modules.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting external data (via WebFetch and mcp__context7) and using it to formulate architecture designs and optimization strategies.
Ingestion points: External documentation retrieved via mcp__context7__get-library-docs and arbitrary URLs via WebFetch.
Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within the retrieved documentation.
Capability inventory: The agent has access to Bash, Write, Edit, and WebFetch, which could be exploited if malicious instructions were successfully injected through the documentation sources.
Sanitization: No input validation or sanitization logic is present for the data retrieved from external tools.

moai-baas-supabase-ext