The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or associated scripts.
[EXTERNAL_DOWNLOADS]: The skill recommends standard integration tools for the Neon platform, including the @neondatabase/serverless driver and the neonctl CLI. These are official resources for the platform and are presented in a neutral, informative context for development purposes.
[COMMAND_EXECUTION]: The scripts/verify.sh utility performs static analysis on local project files to identify potential performance issues or resource leaks (e.g., module-scope connection pooling). The script uses standard system tools like find and grep in a read-only capacity and lacks network or persistence capabilities.
[INDIRECT_PROMPT_INJECTION]: The skill includes an attack surface through its ability to ingest and analyze user-controlled source code files.
Ingestion points: Source code and environment files (.js, .ts, .env) processed by scripts/verify.sh.
Boundary markers: Not applicable; the script uses grep-based pattern matching rather than complex parsing.
Capability inventory: Limited to local filesystem reads and console output within the shell environment.
Sanitization: The script performs literal matches and does not execute or evaluate the content of the scanned files, mitigating common injection vectors.

neon