CLI and MCP server for discovering and calling Google APIs from agents and scripts.
Command
When --input is set, applies the profile to that file (optionally comparing against --golden). When --input is omitted, runs every [[tests]] fixture in the profile file through the expression pipeline and prints a ProfileFixtureResult[] JSON envelope (--format=json).
Command
Parse an expression-profile DSL file and report any errors. Use this in CI to catch malformed catalog profiles before release.
Command
Command "gum-profile" from ehmo/gum, covering gum profile, usage, parent, subcommands and flags.
Command
Command "gum-read" from ehmo/gum, covering gum read, usage, parent, arguments and flags.
Command
Print a JSON description of the active gum command tree, including command paths, aliases, arguments, and flags.
Command
BM25 search the embedded catalog (TTY table, pipe JSON).
Command
Guided setup for a local gum install. It writes agent skills and MCP config for Codex, Claude, Cursor, or Gemini, then prints the Google OAuth and doctor commands needed for first success.
Command
Command "gum-skills-export" from ehmo/gum, covering gum skills export, usage, parent, arguments and flags.
Command
Install gum skills for Codex-compatible agents.
Command
Command "gum-skills-list" from ehmo/gum, covering gum skills list, usage, parent, flags and see also.
Command
Command "gum-skills-show" from ehmo/gum, covering gum skills show, usage, parent, arguments and flags.
Command
List, print, export, or install version-matched agent skills.
Command
Command "gum-version" from ehmo/gum, covering gum version, usage, parent, flags and see also.
Command
Invoke a write-class catalog op. --allow-write is required for the policy gate to admit the dispatch.
Command
Command "gum" from ehmo/gum, covering gum, usage, subcommands, flags and see also.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: