The Agent Skills Directory

[Command Execution]: The skill utilizes a local Node.js script (openai-platform-api-key.mjs) to manage the preparation and decryption of API keys. This script is used to generate public/private key pairs for secure transmission and to write credentials to the local filesystem. This is a core component of the skill's security model, ensuring that plaintext keys are not handled directly by the AI model's text generation process.
[Credential Management & Protection]: A central feature of this skill is its strict enforcement of credential safety. It explicitly forbids commands that would display secrets (such as cat or grep on environment files) and instead uses silent exit-status checks. It also utilizes a dedicated Platform connector for key provisioning, which leverages RSA/JWK encryption to keep secrets out of the tool's standard output.
[File System Interaction]: The skill is designed to interact with sensitive local files, specifically environment configuration files (e.g., .env, .env.local). It includes logic to prefer untracked files and requires explicit user confirmation before writing to files tracked by version control systems, helping to prevent the accidental commit of secrets to repositories.
[Least Privilege & User Consent]: The skill requires a 'Mandatory First Step' where it must inspect for existing credentials and ask for user permission before proceeding with any implementation or execution tasks. This ensures the user maintains control over when and how their API keys are used or generated.

openai-platform-api-key