secrets

Securely manage API keys by storing encrypted environment variable bundles using native OS keychain backends. This prevents exposing credentials in plaintext files or shell history while enabling agents to programmatically access only the specific secrets they need for authentication.

Yes, this secrets management tool works across macOS Keychain, Linux libsecret, and Windows Credential Manager. It facilitates scoped credential injection for CLI-based AI agents across these environments, ensuring maximum security at rest using native OS keychain backends.

Inject specific environment variables by creating a secure bundle and passing it to specific agent runs at runtime. This scoped injection ensures agents only access the credentials required for their task, keeping other bundled credentials hidden from the environment.

Yes, you can use biometric unlocking and session-based caching for agent credentials. The skill implements programmatic read and write access with support for dynamic references, allowing secure access to encrypted system keychains during an active session.

The best way to pass private API keys to a coding agent safely is creating a secure credential bundle and injecting it at runtime. This prevents exposing the keys in your environment variables or shell history while granting the agent the required access.

Storing agent credentials in plaintext environment variables poses a security risk because they can be exposed through shell history or process listing. Using encrypted OS keychain backends with programmatic read and write access secures credentials at rest.