Frequently asked questions
Everything you need to know about Usable.
Usable stores your team's knowledge as workspaces, fragments, files, and collections. Approved AI tools can then search and retrieve that memory when they need trusted context.
You can get started with Usable in under 5 minutes. Simply sign up, connect your tools, and start capturing knowledge. No complex configuration required.
MCP (Model Context Protocol) is an open standard that lets AI tools like Claude Code, Cursor, Cline, VS Code, and Windsurf connect to external data sources. Usable exposes your workspaces over MCP so any compatible AI agent can search and retrieve approved memory during its work.
A memory fragment is a single piece of knowledge stored in Usable — a procedure, decision, runbook, code pattern, or solved problem. Fragments have a type (knowledge, recipe, solution, template, instruction set), tags, relationships, and full text. AI tools retrieve fragments instead of trying to fit everything into a prompt.
Knowledge creation tokens cover putting knowledge in: creating fragments, editing content, summarising source material, and updating workspace memory. Knowledge search tokens cover getting knowledge out: finding fragments, retrieving context, and answering from approved workspaces.
Free: Community support via Discord.
Standard: Email support with priority response times.
Pro: Dedicated Slack channel.
AI agents connect to Usable through Usable Chat, the MCP protocol, or the REST API. When an agent needs context — a runbook, a prior decision, a coding standard — it queries the workspace it's been granted access to, gets the relevant fragments back, and uses them to ground its response. The agent's organisation controls which workspaces each agent can see.
Yes. Usable works with Claude Code, ChatGPT, Cursor, Cline, Windsurf, VS Code, Gemini CLI, Raycast, and any other MCP-compatible tool. You can also integrate via the REST API for tools that don't speak MCP, including internal apps and custom agents.