Glossary
Plain-language definitions for every term on this platform. No jargon left behind.
- OAuth3
- How you give permission to apps. OAuth3 lets you control exactly what an AI agent can do on your behalf, for how long, and revoke access at any time.
- Recipe
- A saved automation you can replay. Recipes record what the AI did so it can repeat the same task without calling an LLM again. This makes repeated tasks nearly free.
- Evidence
- Proof of what happened. Every action the AI takes is recorded with a tamper-evident hash chain, so you can always verify what was done, when, and by whom.
- BYOK (Bring Your Own Key)
- Use your own AI key. You provide your own API key from Anthropic, OpenAI, or other providers. We never mark up the cost and your key stays on your device.
- Rung
- Your trust level. As you use the platform and verify your work, you climb rungs. Higher rungs unlock more capabilities like publishing skills to the store.
- Belt
- Your experience tier. Belts (White, Yellow, Orange, etc.) represent how far you have progressed. They are earned through real usage, not purchased.
- Cloud Twin
- A browser session running in the cloud on your behalf. Your cloud twin can execute recipes even when your computer is off, using the permissions you delegated via OAuth3.
- Skill
- An AI capability packaged for reuse. Skills are versioned, tested, and shared through the Stillwater Store. Think of them as apps for your AI agent.
- Software 5.0
- The platform paradigm. Software 5.0 means AI that produces evidence of what it did, gets cheaper over time, and always keeps humans in control.
- FDA Part 11
- A compliance standard for electronic records. Solace is architected so that every AI action can be audited, making it suitable for regulated industries like healthcare and finance.
- Managed LLM
- AI model access without your own API key. We route your requests to leading AI providers. You pay a small flat fee instead of managing API keys yourself.