The `tea` CLI covers the full everyday workflow: run a script, build a binary, format your files, and run tests. Most users only need a few commands to stay productive. ```bash tea hello.tea tea build hello.tea tea fmt . tea test ``` Start by running a source file directly. When you want a standalone executable, switch to `tea build`. The formatter and test runner use the same command shape, so the CLI stays consistent as projects grow. ## Look up documentation Documentation is bundled with the installed compiler and works offline, from any directory. Start with the language primer, browse a module, or look up one function: ```bash tea docs tea docs language tea docs std.args tea docs std.args.parse_with tea docs @println tea docs --search "command-line" ``` Output is plain Markdown. The documentation landing page also links to the web agent index and Markdown quick reference. ## Run a script Running a Tea file directly is the fastest feedback loop. Pass the script path first, then any arguments you want to forward into the program. ```bash tea hello.tea tea examples/echo/main.tea hello world ``` ## Build a binary Use `tea build` when you want a native executable you can run later without pointing back to the source file. By default, Tea writes the result under `bin/`. ```bash tea build hello.tea ./bin/hello tea build app.tea --output ./dist/app ``` ## Format source files The formatter rewrites Tea files in place. Run it on the current directory for a whole project, or point it at a specific file when you only want to clean up one source. ```bash tea fmt . tea fmt src/main.tea tea fmt . --check ``` ## Run tests The test runner discovers Tea `test` blocks from the paths you give it. You can run a whole test directory, list discovered tests, or narrow execution with a filter. ```bash tea test tea test tests/ tea test --list tea test --filter parser ```