The Agent Skills Directory

[SAFE]: No malicious patterns, prompt injections, or security risks were detected across the skill files or scripts.
[COMMAND_EXECUTION]: The skill includes Python utilities for development workflows (load testing, code generation, and database schema analysis). These scripts operate on user-provided local files or specified URLs as intended and do not perform unauthorized system modifications.
[DATA_EXPOSURE]: No hardcoded secrets or sensitive data exposure patterns were found. The skill uses standard placeholders (e.g., 'token123', 'YOUR_API_KEY_HERE') in documentation examples and correctly instructs users to manage secrets via environment variables.
[INDIRECT_PROMPT_INJECTION]: The api_scaffolder.py and database_migration_tool.py scripts ingest external data formats (OpenAPI specs and SQL DDL). The implementation uses safe parsing methods, such as yaml.safe_load, and limits operations to local file generation, presenting no significant vulnerability surface.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns, such as piping network content to a shell, were detected. Network operations in the load tester are restricted to standard HTTP requests for performance measurement.

senior-backend