Command-line reference

Install status, login, workspace selection, vault operations, local media fetch, search, and MCP client setup.

Install

Authenticate, choose a workspace, manage vault content, search knowledge, and connect MCP-compatible clients from the terminal.

  • Homebrew (macOS / Linux) — brew install liquidlm/tap/liquidlm
  • Install script (macOS / Linux / WSL) — curl -fsSL https://liquidlm.com/install.sh | sh
  • npm (macOS / Linux / Windows) — npm install -g @liquidlm/cli

Sign in and inspect your session

Browser login stores a refreshable local session in ~/.liquidlm. Use whoami to confirm the active organization, bearer class, scopes, and visible vaults.

Switch workspaces explicitly when your account belongs to more than one organization.

liquidlm login
liquidlm whoami
liquidlm org list
liquidlm org use "Workspace name"
liquidlm org show

Vaults, uploads, and search

Vault commands accept an ID or label where noted. Upload accepts files or directories and recursively skips hidden files.

Hybrid semantic plus full-text retrieval is the default. Add --fulltext for lexical-only search or --json for machine-readable search output.

liquidlm vaults list
liquidlm vaults add ./research --label "Research"
liquidlm upload Research ./paper.pdf ./notes/
liquidlm search "What changed in the latest policy?"

Fetch media URLs locally

For hosts that block server-side downloads (notably YouTube), run yt-dlp on your machine. fetch always creates a note with the source URL and captions when available; pass --audio or --video to upload media into the vault.

Requires yt-dlp on PATH (and ffmpeg for --video). Use --cookies-from-browser or --cookies when the site asks you to sign in.

liquidlm fetch Research 'https://youtu.be/…' --notes 'standup clip'
liquidlm fetch Research 'https://youtu.be/…' --audio --cookies-from-browser chrome

Native OAuth — default

Configure the remote endpoint in your client and authorize in the browser. This is always the first choice. Cursor editor and CLI share their configuration; the example below uses native OAuth. See your client's guide for its exact setup.

liquidlm mcp install --client cursor
agent mcp login liquidlm

CLI-managed OAuth — second choice

When native OAuth is unsuitable, log in to LiquidLM and configure the client to launch liquidlm mcp serve over stdio. The CLI refreshes the session without requiring a PAT. Initial authorization requires interaction; log in on the machine running the agent. Use CLI 0.1.5 or newer.

  • Cursor, Codex, Claude Code, Claude Desktop and OpenCode have dedicated guides. ChatGPT web and Claude web cannot launch local processes; use remote OAuth.
unset LIQUIDLM_API_TOKEN LIQUIDLM_TOKEN
liquidlm login
liquidlm whoami
# Configure your client to launch this command:
liquidlm mcp serve

PAT — last fallback

Use a static API token only if both OAuth options are unsuitable. Create a PAT limited to the required actions and vaults, then inject LIQUIDLM_API_TOKEN through your secret manager. Keep its value out of configuration files and shell history.

  • LIQUIDLM_TOKEN is an alias. Environment tokens override the saved session: unset both variables to return to OAuth. Remote clients use different environment-reference syntax; follow the client-specific examples.
# LIQUIDLM_API_TOKEN is already injected into this process.
liquidlm mcp serve

Agent instructions

Generate the current eight-tool LiquidLM instructions in a project without embedding a secret.

liquidlm skill

Current beta boundaries

The CLI covers login, identity, workspace selection, vault management, upload, local media fetch, search, token export, MCP client setup, and skill generation. It is not a multi-machine sync tool or an enterprise administration surface.

References