feat: add non-interactive installation support

[wladimiiir]

Summary

• Add command-line flags to support non-interactive installation for CI/CD pipelines and automated deployments
• When flags are provided, prompts are skipped; missing values gracefully fall back to interactive prompts
• Support --tools none to explicitly skip tool configuration

New CLI Options

Flag	Description
--directory <path>	Installation directory
--modules <modules>	Comma-separated module IDs (e.g., "bmm,bmb")
--tools <tools>	Tool/IDE IDs (use "none" to skip)
--custom-content <paths>	Custom module paths
--action <type>	Action type for existing installations
--user-name <name>	Name for agents to use
--communication-language <lang>	Agent communication language
--document-output-language <lang>	Document output language
--output-folder <path>	Output folder path
-y, --yes	Accept all defaults and skip prompts

Example Usage

# Fully non-interactive
npx bmad-method install \
  --directory ~/myproject \
  --modules bmm,bmb \
  --tools claude-code \
  --yes

# Skip tool configuration
npx bmad-method install --directory ~/myproject --modules bmm --tools none

# Semi-interactive (prompts for missing values)
npx bmad-method install --directory ~/myproject --modules bmm

Test plan

• Test fresh installation with all flags
• Test installation with --tools none
• Test installation with --yes flag
• Test partial flags (should prompt for missing values)
• Test invalid flag values (should show error)
• Verify interactive mode still works as before

🤖 Generated by AiderDesk

[coderabbitai]

📝 Walkthrough

Walkthrough

This change adds non-interactive installation support to BMAD's CLI, introducing command-line flags for automated workflows. Updates include new documentation, expanded CLI options, and modified prompt logic to respect provided configuration values and skip interactive prompts when options are supplied.

Changes

Cohort / File(s)	Summary
Documentation README.md, docs/non-interactive-installation.md	Added Non-Interactive Installation section to README with example usage; created comprehensive guide documenting installation modes, CLI flags, supported actions, module/tool IDs, CI/CD examples, and troubleshooting.
CLI Command & UI Layer tools/cli/commands/install.js, tools/cli/lib/ui.js	Expanded install command with 11 new CLI flags (directory, modules, tools, custom-content, action, user-name, communication-language, document-output-language, output-folder, yes). Modified UI methods to accept options parameter and conditionally bypass prompts or preset values based on CLI-provided configuration.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• fix: resolve ERR_REQUIRE_ESM for inquirer v9+ (#1197) #1278: Modifies the same CLI install files and promptInstall flow; changes to this PR may overlap with async-loading of inquirer in that PR.
• fix(cli): replace inquirer with @clack/prompts for Windows compatibility #1316: Updates the CLI installation UI prompting layer (inquirer to @clack/prompts); this PR's option-handling logic will interact with those prompt changes.

Suggested reviewers

• bmadcode
• alexeyv

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'feat: add non-interactive installation support' directly and clearly describes the main objective of the changeset, which is to add non-interactive installation capabilities via command-line flags.
Description check	✅ Passed	The description comprehensively explains the feature, documents all new CLI options with a clear table, provides multiple usage examples, and outlines a test plan, all directly related to the changeset.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

Important

Action Needed: IP Allowlist Update

If your organization protects your Git platform with IP whitelisting, please add the new CodeRabbit IP address to your allowlist:

• ✨ 136.113.208.247/32 (new)
• 34.170.211.100/32
• 35.222.179.152/32

Reviews will stop working after February 8, 2026 if the new IP is not added to your allowlist.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 2

🤖 Fix all issues with AI agents

In `@tools/cli/lib/ui.js`:
- Around line 753-761: The code assumes os.userInfo() always succeeds when
computing defaultUsername; wrap the os.userInfo() call in a safe fallback: first
try calling os.userInfo() in a try/catch (or guard with typeof), fall back to
process.env.USER || process.env.USERNAME || 'User' if it throws or is undefined,
then compute defaultUsername by capitalizing the first character of that safe
username and assign it to configCollector.collectedConfig.core.user_name; ensure
the console message and existing keys (existingConfig,
configCollector.collectedConfig.core, defaultUsername) remain unchanged
otherwise.
- Around line 317-325: Wrap the module.yaml reading and parsing in a try/catch
to avoid crashes: around the logic that builds moduleYamlPath, reads moduleYaml
with fs.readFile and calls yaml.parse to produce moduleMeta, catch any errors
and log a graceful warning (using console.log/chalk.yellow) that includes the
customPath and the error message, then continue the loop instead of throwing;
ensure you reference moduleYamlPath, moduleYaml, yaml.parse and moduleMeta when
locating where to add the try/catch.

🧹 Nitpick comments (2)

tools/cli/lib/ui.js (2)

296-340: Code duplication: custom content parsing logic appears twice.

The custom content parsing logic (splitting paths, validating, reading module.yaml, building config) is nearly identical here and in lines 407-448 for the new installation flow. Consider extracting this into a helper method to reduce duplication and ensure consistent behavior.

♻️ Suggested helper extraction

+  /**
+   * Parse custom content paths from command-line option
+   * `@param` {string} customContentOption - Comma-separated paths
+   * `@returns` {Object} Custom content config with paths and module IDs
+   */
+  async parseCustomContentPaths(customContentOption) {
+    const paths = customContentOption
+      .split(',')
+      .map((p) => p.trim())
+      .filter(Boolean);
+    console.log(chalk.cyan('Using custom content from command-line:'), chalk.bold(paths.join(', ')));
+
+    const customPaths = [];
+    const selectedModuleIds = [];
+
+    for (const customPath of paths) {
+      const expandedPath = this.expandUserPath(customPath);
+      const validation = this.validateCustomContentPathSync(expandedPath);
+      if (validation !== true) {
+        console.log(chalk.yellow(`⚠️  Skipping invalid custom content path: ${customPath} - ${validation}`));
+        continue;
+      }
+
+      try {
+        const moduleYamlPath = path.join(expandedPath, 'module.yaml');
+        const moduleYaml = await fs.readFile(moduleYamlPath, 'utf-8');
+        const yaml = require('yaml');
+        const moduleMeta = yaml.parse(moduleYaml);
+
+        if (!moduleMeta.code) {
+          console.log(chalk.yellow(`⚠️  Skipping custom content path: ${customPath} - module.yaml missing 'code' field`));
+          continue;
+        }
+
+        customPaths.push(expandedPath);
+        selectedModuleIds.push(moduleMeta.code);
+      } catch (error) {
+        console.log(chalk.yellow(`⚠️  Skipping custom content path: ${customPath} - ${error.message}`));
+      }
+    }
+
+    return { customPaths, selectedModuleIds };
+  }

---

573-585: Consider validating tool IDs from command-line.

When tools are provided via --tools, invalid IDs are silently accepted. Per the documentation, the system should "warn about invalid tool IDs (but won't fail)". Adding validation here would improve user feedback.

🔧 Optional validation

     } else {
       selectedIdes = options.tools
         .split(',')
         .map((t) => t.trim())
         .filter(Boolean);
+
+      // Validate tool IDs and warn about invalid ones
+      const allValidTools = [...preferredIdes, ...otherIdes].map((ide) => ide.value);
+      const invalidTools = selectedIdes.filter((t) => !allValidTools.includes(t));
+      if (invalidTools.length > 0) {
+        console.log(chalk.yellow(`⚠️  Unknown tool ID(s): ${invalidTools.join(', ')} (will be skipped)`));
+        selectedIdes = selectedIdes.filter((t) => allValidTools.includes(t));
+      }
+
       console.log(chalk.cyan('Using tools from command-line:'), chalk.bold(selectedIdes.join(', ')));
     }

[bmadcode]

can you update this against the latest, would love to get this in - there is a conflict @wladimiiir

[wladimiiir]

@bmadcode
Conflicts resolved!

[bmadcode]

Thank you @wladimiiir this is a great add for the final release! Will run some more testing on it also