The Agent Skills Directory

[SAFE]: No external network operations (curl, wget, fetch) or remote data fetching were detected in any of the 84 analyzed files.- [SAFE]: Data processing is performed using standard Python modules (json, csv, re) with no exposure of sensitive file paths or hardcoded credentials.- [SAFE]: The Python automation scripts (generate_prompt.py, batch_generator.py, etc.) are restricted to the Python standard library and do not utilize dangerous functions like eval(), exec(), or subprocess.run().- [SAFE]: No obfuscation patterns, including Base64-encoded instructions, zero-width characters, or homoglyphs, were found in the templates or scripts.- [SAFE]: The skill implements its own safety guardrails via the 'CRITICAL CONSTRAINTS' section in SKILL.md, which strictly limits the agent to prompt generation and prevents unintended task implementation.

prompt-factory