Add CLI improvements, modular commands, and code simplifications#10
Add CLI improvements, modular commands, and code simplifications#10
Conversation
- Bumped version to `2.0.0` to reflect a major iteration. - Simplified `commit-todo-command.js`, replacing unused parameters with stubs for future implementation. - Removed unused backup file `bin/setup.js.backup`. refactor(commands): simplify and stub `commit-todo` and `review` commands - Removed Codex-related logic from `commit-todo-command.js` and `review-command.js` to reduce complexity. - Added placeholders for future implementation of `todo-commit` enforcement workflows. - Updated `listMarkdownFiles` in `file-utils.js` to accept a `baseDir` parameter for improved versatility. - Adjusted imports across modules to reflect the changes. feat(commands): implement `commit-todo` and `review` command handlers - Added `commit-todo` command for enforcing TODO commit policies. - Introduced `review` command for analyzing codebase compliance with architecture and rules. - Modularized commands in `lib/commands` for centralized management. - Enhanced `.dev/` folder setup with templates and automation in `dev-workspace.js`. - Enabled Codex manifest and context index generation across commands. feat(cli): add CLI parser, migration handler, file utilities, and user prompts - Introduced `cli-parser.js` to handle command and option parsing. - Added `config-migrator.js` for configuration migration tasks. - Developed `file-utils.js` for common file system operations with reusable utilities. - Created `prompts.js` module for user interaction during configuration and setup. - Included comprehensive tests for the `cli-parser` module, utilizing AAA principles. refactor: create provider system with template method pattern (Phase 2)
WalkthroughThis pull request introduces a major refactoring of the CLI infrastructure, extracting monolithic command handling into a modular, provider-based architecture. The changes include a new centralized CLI parser, command routing system, reusable command handlers, a provider factory pattern for AI tool setup (Claude, Cursor, Gemini), and comprehensive utilities for file management, templating, user prompts, and workspace configuration. Version bumped to 2.00.0. Changes
Sequence Diagram(s)sequenceDiagram
actor User
participant bin/setup.js as CLI Entry
participant cli-parser.js as Parser
participant Command Handler
participant Provider/Utility
User->>CLI Entry: npm run setup [args]
CLI Entry->>Parser: parseArguments(argv)
Parser-->>CLI Entry: { command, flags, options }
CLI Entry->>Parser: getCommandType(command)
Parser-->>CLI Entry: normalised commandType
alt commandType === 'setup'
CLI Entry->>Command Handler: executeSetup(projectRoot, options)
activate Command Handler
Command Handler->>Provider/Utility: detectAndConfirmLanguage()
Provider/Utility-->>Command Handler: language
Command Handler->>Provider/Utility: selectAITools()
Provider/Utility-->>Command Handler: tools[]
loop for each tool
Command Handler->>Provider/Utility: setupTool(toolName, ...)
Provider/Utility->>Provider/Utility: provider.setup()
note over Provider/Utility: Execute hooks, commands,<br/>settings, additional
Provider/Utility-->>Command Handler: { success }
end
Command Handler->>Provider/Utility: setupDevFolder(...)
Command Handler->>Provider/Utility: setupCentralizedRules(...)
Command Handler->>Provider/Utility: writeCodexManifestAndIndex(...)
Command Handler->>Provider/Utility: setupCodexGuide(...)
Command Handler-->>CLI Entry: (success)
deactivate Command Handler
else commandType === 'update'
CLI Entry->>Command Handler: executeUpdate(projectRoot, options)
Command Handler->>Command Handler: delegates to executeSetup
else commandType === 'review'
CLI Entry->>Command Handler: executeReview(projectRoot, options)
Command Handler->>Provider/Utility: refresh Codex guide
Command Handler->>Provider/Utility: run CodeReviewer.analyze()
Command Handler-->>CLI Entry: exit(1 if violations)
else commandType === 'unknown'
CLI Entry->>CLI Entry: log error, exit(1)
end
CLI Entry-->>User: process exit
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60–90 minutes Areas requiring extra attention:
Poem
Pre-merge checks and finishing touches✅ Passed checks (3 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
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. Comment |
Greptile OverviewGreptile SummaryThis PR refactors the CLI from a monolithic 1700+ line file into a modular architecture with separated concerns, introducing command handlers, a provider system using the Template Method pattern, and comprehensive utility modules. Major Changes:
Code Quality:
Issues Found:
Confidence Score: 3/5
Sequence DiagramsequenceDiagram
participant User
participant CLI as bin/setup.js
participant Parser as cli-parser
participant SetupCmd as setup-command
participant Prompts as prompts
participant ProviderFactory as providers/index
participant Provider as BaseProvider/Subclass
participant DevWorkspace as dev-workspace
participant TemplateManager as template-manager
User->>CLI: Execute ai-dotfiles-manager
CLI->>Parser: parseArguments(argv)
Parser-->>CLI: {command, options}
CLI->>Parser: getCommandType(command)
Parser-->>CLI: 'setup'
CLI->>SetupCmd: executeSetup(projectRoot, options)
SetupCmd->>SetupCmd: detectAndConfirmLanguage()
alt autoYes mode
SetupCmd->>SetupCmd: Use detected/default language
else interactive mode
SetupCmd->>Prompts: confirmLanguage(detectedLanguage)
Prompts-->>SetupCmd: confirmed language
end
SetupCmd->>SetupCmd: selectAITools(autoYes)
alt autoYes mode
SetupCmd->>SetupCmd: Default to ['claude']
else interactive mode
SetupCmd->>Prompts: selectTools()
Prompts-->>SetupCmd: ['claude', 'gemini', ...]
end
loop For each selected tool
SetupCmd->>ProviderFactory: setupTool(tool, projectRoot, templatesDir, options)
ProviderFactory->>ProviderFactory: createProvider(tool, ...)
ProviderFactory->>Provider: new Provider(...)
Provider-->>ProviderFactory: provider instance
ProviderFactory->>Provider: provider.setup()
Provider->>Provider: Check for existing config
alt Existing config found
Provider->>Prompts: promptMigrationAction()
Prompts-->>Provider: migration action
Provider->>Provider: handleExistingConfig(action)
end
Provider->>Provider: setupHooks()
Provider->>Provider: setupCommands()
Provider->>Provider: setupSettings()
Provider->>Provider: setupAdditional()
Provider-->>ProviderFactory: {success: true}
ProviderFactory-->>SetupCmd: setup result
end
SetupCmd->>DevWorkspace: setupDevFolder(projectRoot, language, false, {})
DevWorkspace->>TemplateManager: generateArchitectureDoc()
TemplateManager-->>DevWorkspace: architecture content
DevWorkspace-->>SetupCmd: done
SetupCmd->>DevWorkspace: setupCentralizedRules(projectRoot, language, ...)
DevWorkspace-->>SetupCmd: done
alt !noCodexGuide
SetupCmd->>DevWorkspace: ensureAgentsTemplate(projectRoot)
SetupCmd->>TemplateManager: discoverRuleFiles(projectRoot, language)
TemplateManager-->>SetupCmd: discovered rules
SetupCmd->>DevWorkspace: writeCodexManifestAndIndex(...)
SetupCmd->>DevWorkspace: setupCodexGuide(...)
end
alt language === 'typescript'
SetupCmd->>SetupCmd: setupTypeScriptConfig(projectRoot, false, templatesDir)
end
SetupCmd->>SetupCmd: printNextSteps(tools, language, false)
SetupCmd-->>CLI: done
CLI-->>User: Success message
|
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
| { | ||
| "name": "ai-dotfiles-manager", | ||
| "version": "1.14.0", | ||
| "version": "2.00.0", |
There was a problem hiding this comment.
syntax: version format should be 2.0.0 not 2.00.0 (semantic versioning format)
| "version": "2.00.0", | |
| "version": "2.0.0", |
Prompt To Fix With AI
This is a comment left during a code review.
Path: package.json
Line: 3:3
Comment:
**syntax:** version format should be `2.0.0` not `2.00.0` (semantic versioning format)
```suggestion
"version": "2.0.0",
```
How can I resolve this? If you propose a fix, please make it concise.| function createProvider(toolName, projectRoot, templatesDir, options = {}) { | ||
| switch (toolName) { | ||
| case 'claude': | ||
| return new ClaudeProvider(projectRoot, templatesDir, options); | ||
| case 'gemini': | ||
| return new GeminiProvider(projectRoot, templatesDir, options); | ||
| case 'cursor': | ||
| return new CursorProvider(projectRoot, templatesDir, options); | ||
| default: | ||
| throw new Error(`Unknown tool: ${toolName}`); | ||
| } |
There was a problem hiding this comment.
logic: createProvider will throw error for kilo and roo tools that are selectable in prompts (lib/prompts.js:65-66), causing runtime crash if user selects them
Prompt To Fix With AI
This is a comment left during a code review.
Path: lib/providers/index.js
Line: 18:28
Comment:
**logic:** `createProvider` will throw error for `kilo` and `roo` tools that are selectable in prompts (`lib/prompts.js:65-66`), causing runtime crash if user selects them
How can I resolve this? If you propose a fix, please make it concise.![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| async function selectTools(inquirerModule = inquirer) { | ||
| const { tools } = await inquirerModule.prompt([ | ||
| { | ||
| type: 'checkbox', | ||
| name: 'tools', | ||
| message: 'Which AI tools would you like to configure? (Space to select)', | ||
| choices: [ | ||
| { name: '✨ Select All', value: 'all', checked: true }, | ||
| new inquirerModule.Separator(), | ||
| { name: 'Claude Code', value: 'claude', checked: true }, | ||
| { name: 'Gemini CLI', value: 'gemini', checked: true }, | ||
| { name: 'Cursor', value: 'cursor', checked: false }, | ||
| { name: 'Kilo Code', value: 'kilo', checked: false }, | ||
| { name: 'Roo Code', value: 'roo', checked: false }, | ||
| new inquirerModule.Separator(), | ||
| { name: '🚫 None (Codex only — skip provider folders)', value: 'none', checked: false }, | ||
| ], | ||
| }, | ||
| ]); | ||
|
|
||
| // Handle "Select All" option | ||
| if (tools.includes('all')) { | ||
| return ['claude', 'gemini', 'cursor', 'kilo', 'roo']; | ||
| } |
There was a problem hiding this comment.
The tool-selection prompt advertises Kilo Code and Roo Code and returns them whenever the default “Select All” option remains checked, but the provider factory only knows about claude, gemini, and cursor. Choosing the defaults (or explicitly picking Kilo/Roo) causes setupTool to throw Unknown tool: kilo/roo and abort the setup process before any configuration is applied. Users running ai-dotfiles-manager setup interactively will hit this failure path immediately. Either implement providers for these entries or remove them from the prompt until they are supported.
Useful? React with 👍 / 👎.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
🧹 Nitpick comments (8)
lib/commands/review-command.js (1)
22-30: Log Codex refresh failures.
Silent failure hides actionable feedback when detectLanguage/discoverRuleFiles/setupCodexGuide throw. Emitting a warning (e.g.,console.warn) keeps the command non-fatal while still surfacing the issue to the operator.lib/providers/cursor-provider.js (2)
13-28: Consider extending BaseProvider for architectural consistency.Unlike
ClaudeProviderandGeminiProvider,CursorProviderdoesn't extendBaseProvider. This creates architectural inconsistency and duplicates DI patterns (fs, log) that are already provided by the base class.Consider refactoring to extend
BaseProvider:-class CursorProvider { +class CursorProvider extends BaseProvider { constructor(projectRoot, templatesDir, options = {}) { - this.projectRoot = projectRoot; - this.templatesDir = templatesDir; - this.autoYes = options.autoYes || false; - this.fs = options.fs || fs; - this.log = options.log || console.log; + super('cursor', 'Cursor', projectRoot, templatesDir, options); this.cursorRulesPath = path.join(projectRoot, '.cursorrules'); - this.templatePath = path.join(templatesDir, 'cursor', '.cursorrules'); + this.templatePath = path.join(this.templateDir, '.cursorrules'); }
56-62: Consider externalizing the local rules template content.The
.cursorrules.localcontent is hardcoded. For consistency with the template-driven approach used elsewhere, consider storing this content in a template file.lib/file-utils.js (1)
16-31: Consider adding error handling for robustness.The function lacks error handling for file system operations (
mkdirSync,readdirSync,statSync,copyFileSync). WhilstcopyPathwraps this in a try-catch, direct callers of this function would encounter unhandled exceptions.Consider either:
- Adding internal error handling with informative error messages, or
- Documenting that this function may throw and should be wrapped by callers
lib/commands/setup-command.js (4)
12-13: Combine imports from the same module.Lines 12-13 import from
../template-managerseparately. Consider combining them into a single destructuring statement for better readability.Apply this diff:
-const { discoverRuleFiles } = require('../template-manager'); -const { getPackageRoot, getTemplatesDir } = require('../template-manager'); +const { discoverRuleFiles, getPackageRoot, getTemplatesDir } = require('../template-manager');
23-73: Consider adding error handling.The
executeSetupfunction performs multiple async operations (file I/O, user prompts, tool setup) without try-catch error handling. If any operation fails, the error will propagate unhandled, potentially resulting in a poor user experience with cryptic error messages.Consider wrapping the setup flow in a try-catch block:
async function executeSetup(projectRoot, options = {}) { const autoYes = options.autoYes || false; const noCodexGuide = options.noCodexGuide || false; console.log(chalk.blue.bold('\n🤖 AI Dotfiles Manager Setup\n')); if (autoYes) { console.log(chalk.gray('Running in non-interactive mode (--yes flag)\n')); } + try { // Step 1: Detect and confirm language const language = await detectAndConfirmLanguage(projectRoot, autoYes); // ... rest of the steps ... console.log(chalk.green.bold('\n✅ Setup complete!\n')); printNextSteps(tools, language, false); + } catch (error) { + console.error(chalk.red(`\n❌ Setup failed: ${error.message}\n`)); + throw error; + } }
141-142: Move require statements to the top of the module.The
fsandpathmodules are required inside the function rather than at the module level. This is inconsistent with the rest of the file and considered a code smell.Apply this diff:
const chalk = require('chalk'); +const fs = require('fs'); +const path = require('path'); const { detectLanguage } = require('../language-detector'); // ... rest of imports async function setupTypeScriptConfig(projectRoot, isUpdate, templatesDir) { - const fs = require('fs'); - const path = require('path'); - console.log(chalk.blue('\n📦 Setting up TypeScript configuration files...'));
153-178: Add error handling for file operations.The
fs.copyFileSyncoperations at lines 172 and 176 can throw errors (e.g., permission denied, disk full, source not readable), but there's no error handling. This could result in the setup process failing midway through with an unhelpful error message.Consider wrapping file operations in try-catch blocks:
for (const { source, dest } of tsConfigFiles) { const sourcePath = path.join(templatesDir, 'languages', 'typescript', source); const destPath = path.join(projectRoot, dest); if (!fs.existsSync(sourcePath)) { console.log(chalk.yellow(` ⚠ Template file ${source} not found, skipping`)); continue; } + try { if (fs.existsSync(destPath)) { const backupPath = `${destPath}.bak`; let backupNumber = 1; let finalBackupPath = backupPath; while (fs.existsSync(finalBackupPath)) { finalBackupPath = `${destPath}.bak${backupNumber}`; backupNumber++; } fs.copyFileSync(destPath, finalBackupPath); console.log(chalk.yellow(` ⚠ Backed up existing ${dest} to ${path.basename(finalBackupPath)}`)); } fs.copyFileSync(sourcePath, destPath); console.log(chalk.green(` ✓ Created ${dest}`)); + } catch (error) { + console.log(chalk.red(` ✗ Failed to copy ${dest}: ${error.message}`)); + } }
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Knowledge base: Disabled due to Reviews -> Disable Knowledge Base setting
📒 Files selected for processing (20)
__tests__/lib/cli-parser.test.js(1 hunks)bin/setup.js(2 hunks)lib/cli-parser.js(1 hunks)lib/commands/commit-todo-command.js(1 hunks)lib/commands/index.js(1 hunks)lib/commands/review-command.js(1 hunks)lib/commands/setup-command.js(1 hunks)lib/commands/update-command.js(1 hunks)lib/dev-workspace.js(1 hunks)lib/file-utils.js(1 hunks)lib/migration/config-migrator.js(1 hunks)lib/prompts.js(1 hunks)lib/providers/base-provider.js(1 hunks)lib/providers/claude-provider.js(1 hunks)lib/providers/cursor-provider.js(1 hunks)lib/providers/gemini-provider.js(1 hunks)lib/providers/index.js(1 hunks)lib/template-manager.js(1 hunks)package.json(1 hunks)tsconfig.json.bak(0 hunks)
💤 Files with no reviewable changes (1)
- tsconfig.json.bak
🧰 Additional context used
🪛 Biome (2.1.2)
bin/setup.js
[error] 60-60: Other switch clauses can erroneously access this declaration.
Wrap the declaration in a block to restrict its access to the switch clause.
The declaration is defined in this switch clause:
Safe fix: Wrap the declaration in a block.
(lint/correctness/noSwitchDeclarations)
🔇 Additional comments (25)
package.json (1)
3-3: Version bump looks good.
No concerns with the release metadata update.lib/commands/commit-todo-command.js (1)
17-32: Stub communicates status clearly.
Graceful messaging for the forthcoming todo enforcement keeps users informed without breaking flow.lib/commands/update-command.js (1)
15-18: Delegation to setup is appropriate.
Reusing executeSetup keeps the update path consistent with the primary install flow.__tests__/lib/cli-parser.test.js (1)
8-184: Comprehensive parser coverage.
Thorough assertions over flags, aliases, and fallback behaviour give strong confidence in the new CLI parser.lib/providers/index.js (2)
18-29: LGTM! Clean factory pattern implementation.The factory function correctly instantiates providers based on tool name and handles unknown tools appropriately.
39-42: LGTM! Simple and effective setup orchestration.The function correctly delegates to the provider's setup method. Error handling is appropriately left to the caller.
lib/commands/index.js (1)
1-16: LGTM! Clean module organization.The centralized command handler exports provide a clean API surface for the CLI entry point.
lib/cli-parser.js (1)
39-53: LGTM! Solid command normalization.The function correctly handles command aliases and provides sensible defaults.
lib/providers/gemini-provider.js (2)
19-31: LGTM! Well-structured command setup.The method properly validates template existence and provides clear user feedback.
36-42: LGTM! Clean settings setup.The method correctly delegates to the base provider's template copying functionality.
lib/providers/cursor-provider.js (1)
71-100: LGTM! Robust existing configuration handling.The method properly handles user choices and delegates migration tasks appropriately.
lib/providers/claude-provider.js (3)
22-52: LGTM! Efficient global command setup with content comparison.The method intelligently avoids unnecessary file writes by comparing content before copying.
57-83: LGTM! Conservative and user-friendly settings management.The method appropriately preserves existing project-specific settings in non-interactive mode whilst prompting for confirmation in interactive mode.
88-100: LGTM! Straightforward workflows setup.The method correctly handles optional workflows directory with appropriate validation.
lib/file-utils.js (4)
78-82: LGTM! Simple directory creation utility.The function correctly uses
recursive: trueto create parent directories as needed.
90-94: LGTM! Clean file writing with parent directory creation.The function properly ensures parent directories exist before writing.
103-113: LGTM! Robust markdown file listing with proper error handling.The function correctly handles missing directories and normalises paths to POSIX format.
120-126: LGTM! Cross-platform executable flag handling.The function correctly handles Windows platforms where chmod may fail.
bin/setup.js (3)
14-30: LGTM! Clean argument parsing and flag handling.The centralized CLI parser provides a clean separation of concerns and proper early exit behaviour for version and help flags.
72-79: LGTM! Comprehensive error handling.The error handling properly logs both the error message and stack trace before exiting with a non-zero status code.
84-114: LGTM! Comprehensive and well-organised help documentation.The help message provides clear usage information, examples, and proper formatting.
lib/commands/setup-command.js (4)
81-109: LGTM!The language detection and confirmation logic is well-structured, handling both auto and interactive modes gracefully with appropriate defaults.
116-131: LGTM!The tool selection logic is clear and handles both auto and interactive modes appropriately.
189-246: LGTM!The
printNextStepsfunction provides comprehensive, well-formatted guidance to users based on their configuration choices, enhancing the user experience significantly.
248-254: LGTM!The module exports are appropriate and expose all the necessary functions for external use.
| case 'commit-todo': | ||
| const subCommand = parsed.commandArgs[0] || 'enforce'; | ||
| await executeCommitTodo(PROJECT_ROOT, { | ||
| noCodexGuide: parsed.options.noCodexGuide, | ||
| }, subCommand); | ||
| break; |
There was a problem hiding this comment.
Fix variable declaration scope in switch statement.
The const subCommand declaration at line 60 is accessible from other switch cases, which violates block scoping expectations and creates a potential temporal dead zone (TDZ) issue. This is flagged by the static analysis tool.
Apply this diff to properly scope the declaration:
case 'commit-todo':
+ {
const subCommand = parsed.commandArgs[0] || 'enforce';
await executeCommitTodo(PROJECT_ROOT, {
noCodexGuide: parsed.options.noCodexGuide,
}, subCommand);
break;
+ }📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| case 'commit-todo': | |
| const subCommand = parsed.commandArgs[0] || 'enforce'; | |
| await executeCommitTodo(PROJECT_ROOT, { | |
| noCodexGuide: parsed.options.noCodexGuide, | |
| }, subCommand); | |
| break; | |
| case 'commit-todo': | |
| { | |
| const subCommand = parsed.commandArgs[0] || 'enforce'; | |
| await executeCommitTodo(PROJECT_ROOT, { | |
| noCodexGuide: parsed.options.noCodexGuide, | |
| }, subCommand); | |
| break; | |
| } |
🧰 Tools
🪛 Biome (2.1.2)
[error] 60-60: Other switch clauses can erroneously access this declaration.
Wrap the declaration in a block to restrict its access to the switch clause.
The declaration is defined in this switch clause:
Safe fix: Wrap the declaration in a block.
(lint/correctness/noSwitchDeclarations)
🤖 Prompt for AI Agents
In bin/setup.js around lines 59 to 64, the const subCommand is declared directly
in the switch clause and can leak into other cases causing TDZ/static analysis
issues; fix by introducing a block scope for the 'commit-todo' case (add { }
around the case body) and move the const subCommand declaration and its await
executeCommitTodo call inside that block so the variable is block-scoped to this
case only.
| function parseArguments(argv = []) { | ||
| const flags = argv.filter(arg => arg.startsWith('-')); | ||
| const positionals = argv.filter(arg => !arg.startsWith('-')); | ||
| const command = positionals[0] || null; | ||
| const commandIndex = typeof command === 'string' ? argv.indexOf(command) : -1; | ||
|
|
||
| return { | ||
| command, | ||
| commandArgs: commandIndex >= 0 ? argv.slice(commandIndex + 1) : [], | ||
| flags, | ||
| options: { | ||
| autoYes: flags.includes('--yes') || flags.includes('-y'), | ||
| noCodexGuide: flags.includes('--no-codex-guide'), | ||
| detailed: flags.includes('--detailed'), | ||
| json: flags.includes('--json'), | ||
| fix: flags.includes('--fix'), | ||
| help: flags.includes('--help') || flags.includes('-h'), | ||
| version: flags.includes('--version') || flags.includes('-v'), | ||
| } | ||
| }; | ||
| } |
There was a problem hiding this comment.
Fix commandArgs calculation for edge case when command is null.
When no positional arguments are provided, commandIndex is -1, causing argv.slice(-1 + 1) to return the entire argv array (including flags) instead of an empty array.
Example:
parseArguments(['--yes', '--help'])
// command = null, commandIndex = -1
// commandArgs = argv.slice(0) = ['--yes', '--help'] ❌
// Expected: commandArgs = []Apply this diff to fix the issue:
return {
command,
- commandArgs: commandIndex >= 0 ? argv.slice(commandIndex + 1) : [],
+ commandArgs: command !== null ? argv.slice(commandIndex + 1) : [],
flags,
options: {🤖 Prompt for AI Agents
In lib/cli-parser.js around lines 12 to 32, the commandArgs calculation uses
commandIndex when command is null which yields -1 and makes argv.slice(0) return
all flags; change the logic to only compute/slice when a command exists and was
found: determine commandIndex only if command is non-null and present in argv
(or simply branch on command truthiness) and set commandArgs to an empty array
when there is no command, otherwise use argv.slice(commandIndex + 1) to capture
only arguments after the command.
| function copyPath(source, target, replace = true, fsModule = fs) { | ||
| const isDirectory = fsModule.statSync(source).isDirectory(); | ||
|
|
||
| if (fsModule.existsSync(target)) { | ||
| if (!replace) { | ||
| return { success: false, skipped: true }; | ||
| } | ||
|
|
||
| const stats = fsModule.lstatSync(target); | ||
| if (stats.isDirectory()) { | ||
| fsModule.rmSync(target, { recursive: true }); | ||
| } else { | ||
| fsModule.unlinkSync(target); | ||
| } | ||
| } | ||
|
|
||
| try { | ||
| if (isDirectory) { | ||
| copyDirectorySync(source, target, fsModule); | ||
| } else { | ||
| const parent = path.dirname(target); | ||
| if (!fsModule.existsSync(parent)) { | ||
| fsModule.mkdirSync(parent, { recursive: true }); | ||
| } | ||
| fsModule.copyFileSync(source, target); | ||
| } | ||
| return { success: true }; | ||
| } catch (error) { | ||
| return { success: false, error: error.message }; | ||
| } | ||
| } |
There was a problem hiding this comment.
Potential error: statSync outside try-catch block.
Line 42 calls statSync outside the try-catch block, which could throw an error if the source path doesn't exist. This would result in an unhandled exception rather than the expected { success: false, error } return value.
Apply this diff to fix:
function copyPath(source, target, replace = true, fsModule = fs) {
+ try {
const isDirectory = fsModule.statSync(source).isDirectory();
if (fsModule.existsSync(target)) {
// ... existing code ...
}
- try {
if (isDirectory) {
copyDirectorySync(source, target, fsModule);
} else {
// ... existing code ...
}
return { success: true };
} catch (error) {
return { success: false, error: error.message };
}
}Committable suggestion skipped: line range outside the PR's diff.
🤖 Prompt for AI Agents
In lib/file-utils.js around lines 41 to 71, the call to
fsModule.statSync(source) is made outside the try-catch so a missing/invalid
source throws an unhandled exception; wrap the statSync (and any dependent logic
like isDirectory determination and subsequent file checks) inside the existing
try-catch so errors from statSync are caught and returned as { success: false,
error: error.message }, or alternatively perform a safe exists check before
statSync to early-return an error — ensure all filesystem ops that can throw are
inside try-catch and preserve the original return shapes.
| const { tools } = await inquirerModule.prompt([ | ||
| { | ||
| type: 'checkbox', | ||
| name: 'tools', | ||
| message: 'Which AI tools would you like to configure? (Space to select)', | ||
| choices: [ | ||
| { name: '✨ Select All', value: 'all', checked: true }, | ||
| new inquirerModule.Separator(), | ||
| { name: 'Claude Code', value: 'claude', checked: true }, | ||
| { name: 'Gemini CLI', value: 'gemini', checked: true }, | ||
| { name: 'Cursor', value: 'cursor', checked: false }, | ||
| { name: 'Kilo Code', value: 'kilo', checked: false }, | ||
| { name: 'Roo Code', value: 'roo', checked: false }, | ||
| new inquirerModule.Separator(), | ||
| { name: '🚫 None (Codex only — skip provider folders)', value: 'none', checked: false }, | ||
| ], | ||
| }, | ||
| ]); | ||
|
|
||
| // Handle "Select All" option | ||
| if (tools.includes('all')) { | ||
| return ['claude', 'gemini', 'cursor', 'kilo', 'roo']; | ||
| } | ||
|
|
||
| // Handle "None" option | ||
| if (tools.includes('none')) { | ||
| return []; | ||
| } | ||
|
|
||
| return tools; | ||
| } |
There was a problem hiding this comment.
Stop auto-selecting every provider
With “Select All” pre-checked, the returned selection always contains every provider even when the user unticks some (e.g. Cursor stays unchecked in the UI but is still returned). This contradicts what the checkbox list communicates and prevents opting into a subset, so the CLI will set up unwanted providers. Please make “Select All” opt-in and strip the sentinel values before returning the list.
- { name: '✨ Select All', value: 'all', checked: true },
+ { name: '✨ Select All', value: 'all', checked: false },
@@
- // Handle "Select All" option
- if (tools.includes('all')) {
- return ['claude', 'gemini', 'cursor', 'kilo', 'roo'];
- }
-
- // Handle "None" option
- if (tools.includes('none')) {
- return [];
- }
-
- return tools;
+ // Handle "None" option
+ if (tools.includes('none')) {
+ return [];
+ }
+
+ // Handle "Select All" option
+ if (tools.includes('all')) {
+ return ['claude', 'gemini', 'cursor', 'kilo', 'roo'];
+ }
+
+ return tools.filter(tool => tool !== 'all');1 participant
Summary of Changes
2.0.0, marking a major iteration.commit-todo,review) with stubs for future features.commit-todoandreviewcommands..dev/folder for automation.Checklist
Additional Notes
Summary by CodeRabbit
Release Notes
New Features
Chores