Smart Prompt Builder

This skill appears coherent and low-risk: it only formats prompts from local inputs and a style YAML. Before installing/running, consider these simple precautions: - Inspect the full scripts/build_prompt.py (the provided file was large; ensure the remainder contains no network calls, subprocess.exec/os.system usage, or hidden endpoints that could exfiltrate data). - Run it in a sandbox or virtualenv and review scripts/requirements.txt before pip installing dependencies. - Avoid passing sensitive secrets or private data (passwords, API keys, full documents) into the --context or --corpus-results arguments, since output prompts are intended to be consumed by downstream LLMs and could be logged or transmitted by other tools. - If you plan to integrate with other skills (e.g., corpus-search or novel-writer), review those integrations for network endpoints and credential use. If you want, I can scan the remainder of build_prompt.py for any network/subprocess/file-access patterns you should be aware of.