Operate your memory with explicit commands.
The command line gives operators a direct way to inspect and maintain a deployment. It complements the MCP tools and admin panel with explicit import, export and diagnostic operations.
Illustrative example
How can I check an import before it changes your memory?
Use the supported dry-run or preview mode, review the source and counts, then run the operation with the intended options.
Illustrative scenario, not a live product result.Load and move your knowledge deliberately.
CLI commands support transcript imports, local transcript pushes, Markdown export and preview-first Markdown import. GitHub issue and pull-request import is available as an operator-run command. These are concrete paths for bringing material into memory; they do not imply a continuously connected inbox, calendar or third-party workspace. Choose the smallest representative sample before applying an import to a large archive.
Manage the deployment around the content.
Operators can register sources, issue credentials, set budgets and inspect current access. Maintenance and diagnostic commands help answer whether content is stored, indexed and ready for retrieval. The commands run where the required database or service access exists. A read-only diagnostic is different from a remediation or bulk write, so inspect the command’s help and proposed changes before running it.
Use the installed version as the reference.
The live help output lists the commands and options supported by that checkout. Public guides explain the common workflows, but version differences and deployment configuration still matter. Keep secrets out of command history and use documented token-file or secret-store paths where available. For a new installation, the quickstart covers the infrastructure prerequisites and the first successful client connection.
Keep the context. Move the work forward.
Connect your knowledge to the AI you already use. Find your starting point and make the next conversation count.