codexa 1.2.0 → 1.2.2

package/README.md

@@ -51,7 +51,6 @@
 - 🤖 **Multiple LLM Support**: Works with Groq (cloud)
 - 💾 **Local Storage**: SQLite database for embeddings and context
 - 🎯 **Smart Chunking**: Intelligent code splitting with configurable overlap
-- 🔄 **Session Management**: Maintain conversation context across queries
 - 📊 **Streaming Output**: Real-time response streaming for better UX
 - 🎨 **Multiple File Types**: Supports TypeScript, JavaScript, Python, Go, Rust, Java, and more
 - 🧠 **Smart Configuration**: Automatically detects project languages and optimizes config
@@ -59,6 +58,8 @@
 - ⚙️ **Highly Configurable**: Fine-tune chunking, retrieval, and model parameters
 - 🚀 **Zero Setup**: Works out of the box with sensible defaults
 
+> ⚠️ **Codebase Size Limitation**: Codexa is optimized for small to medium-sized codebases. It currently supports projects with up to **200 files** and **20,000 chunks**. For larger codebases, consider using more restrictive `includeGlobs` patterns to focus on specific directories or file types.
+
 ## Installation
 
 ### Prerequisites
@@ -87,7 +88,7 @@ npm install -g codexa
 Verify installation:
 
 ```bash
-codexa --version
+codexa
 ```
 
 #### Method 2: Homebrew (macOS)
@@ -143,35 +144,20 @@ Groq provides fast cloud-based LLMs with a generous free tier.
 4. Create a new API key
 5. Copy your API key (starts with `gsk_`)
 
-**Step 2: Set Environment Variable**
-
-**macOS/Linux:**
-```bash
-# Add to your shell profile (~/.zshrc, ~/.bashrc, etc.)
-export GROQ_API_KEY="gsk_your_api_key_here"
-
-# Reload your shell or run:
-source ~/.zshrc  # or ~/.bashrc
-```
+**Step 2: Set GROQ API Key**
 
-**Windows (PowerShell):**
-```powershell
-$env:GROQ_API_KEY="gsk_your_api_key_here"
+Run the following command to securely save your API key:
 
-# Or add permanently:
-[System.Environment]::SetEnvironmentVariable('GROQ_API_KEY', 'gsk_your_api_key_here', 'User')
+```bash
+codexa config set GROQ_API_KEY "gsk_your_api_key_here"
 ```
 
-**Windows (Command Prompt):**
-```cmd
-setx GROQ_API_KEY "gsk_your_api_key_here"
-```
+This will save the key to your local configuration file (`.codexarc.json`).
 
 **Step 3: Verify API Key is Set**
 
 ```bash
-echo $GROQ_API_KEY  # macOS/Linux
-echo %GROQ_API_KEY% # Windows CMD
+codexa config get GROQ_API_KEY
 ```
 
 **Step 4: Configure Codexa**
@@ -181,15 +167,15 @@ Codexa defaults to using Groq when you run `codexa init`. If you need to manuall
 ```json
 {
   "modelProvider": "groq",
-  "model": "llama-3.1-8b-instant",
+  "model": "openai/gpt-oss-120b",
   "embeddingProvider": "local",
   "embeddingModel": "Xenova/all-MiniLM-L6-v2"
 }
 ```
 
-**Available Groq Models:**
-- `llama-3.1-8b-instant` - Fast responses (recommended, default)
-- `llama-3.1-70b-versatile` - Higher quality, slower
+**Models you can use:**
+- `openai/gpt-oss-120b` (recommended, default)
+- `llama-3.1-70b-versatile`
 
 
 
@@ -198,13 +184,14 @@ Codexa defaults to using Groq when you run `codexa init`. If you need to manuall
 **For Groq:**
 ```bash
 # 1. Get API key from console.groq.com
-# 2. Set environment variable
-export GROQ_API_KEY="gsk_your_key"
 
-# 3. Run codexa init (defaults to Groq)
+# 2. Run codexa init (defaults to Groq)
 codexa init
 
-# 4. Ready to use!
+# 3. Set GROQ API key
+codexa config set GROQ_API_KEY "gsk_your_key"
+
+# 4. Proceed to igestion
 ```
 
 
@@ -224,13 +211,19 @@ Once Codexa is installed and your LLM is configured, you're ready to use it:
    ```
    This creates a `.codexarc.json` configuration file with sensible defaults.
 
-3. **Ingest your codebase:**
+3. **Set GROQ API Key**
+   ```bash
+   codexa config set GROQ_API_KEY "gsk_your_key"
+   ```
+   This will save the key to your local configuration file (`.codexarc.json`).
+
+4. **Ingest your codebase:**
    ```bash
    codexa ingest
    ```
    This indexes your codebase and creates embeddings. First run may take a few minutes.
 
-4. **Ask questions:**
+5. **Ask questions:**
    ```bash
    codexa ask "How does the authentication flow work?"
    codexa ask "What is the main entry point of this application?"
@@ -269,9 +262,10 @@ Analyzing codebase...
 │                                                             │
 │   Next Steps:                                               │
 │                                                             │
-│   1. Review .codexarc.json - Update provider keys if needed │
-│   2. Run: codexa ingest - Start indexing your codebase      │
-│   3. Run: codexa ask "your question" - Ask questions        │
+│   1. Review .codexarc.json - Update provider keys if needed |
+│   2. Set your GROQ API Key: codexa config set GROQ_API_KEY  |
+│   3. Run: codexa ingest - Start indexing your codebase      │
+│   4. Run: codexa ask "your question" - Ask questions        │
 │                                                             │
 └─────────────────────────────────────────────────────────────┘
 ```
@@ -314,6 +308,30 @@ codexa ingest --force
 
 ---
 
+### `config`
+
+Manage configuration values, including API keys.
+
+```bash
+codexa config <action> [key] [value]
+```
+
+**Actions:**
+- `set <key> <value>` - Set a configuration value
+- `get <key>` - Get a configuration value
+- `list` - List all configuration values
+
+**Examples:**
+```bash
+# Set Groq API Key
+codexa config set GROQ_API_KEY "gsk_..."
+
+# Check current key
+codexa config get GROQ_API_KEY
+```
+
+---
+
 ### `ask`
 
 Ask natural language questions about your codebase.
@@ -326,8 +344,8 @@ codexa ask <question...> [options]
 - `<question...>` - Your question (can be multiple words)
 
 **Options:**
-- `-s, --session <name>` - Session identifier to maintain conversation context (default: `"default"`)
-- `--no-stream` - Disable streaming output (show full response at once)
+<!-- - `-s, --session <name>` - Session identifier to maintain conversation context (default: `"default"`) -->
+- `--stream` - Enable streaming output
 
 **Examples:**
 ```bash
@@ -337,14 +355,8 @@ codexa ask "How does user authentication work?"
 # Question with multiple words
 codexa ask "What is the main entry point of this application?"
 
-# Use a specific session for context
-codexa ask "How does the login function work?" --session my-analysis
-
-# Disable streaming
-codexa ask "Summarize the codebase structure" --no-stream
-
-# Follow-up question in the same session
-codexa ask "Can you explain that in more detail?" --session my-analysis
+# Enable streaming
+codexa ask "Summarize the codebase structure" --stream
 ```
 
 **How it works:**
@@ -393,12 +405,15 @@ Some settings can be configured via environment variables:
 | Variable | Description | Required For |
 |----------|-------------|--------------|
 | `GROQ_API_KEY` | Groq API key for cloud LLM | Groq provider |
-| `OPENAI_API_KEY` | OpenAI API key (for embeddings) | OpenAI embeddings |
+<!-- | `OPENAI_API_KEY` | OpenAI API key (for embeddings) | OpenAI embeddings | -->
 
 **Example:**
 ```bash
-export GROQ_API_KEY="gsk_your_key_here"
-export OPENAI_API_KEY="sk-your_key_here"  # If using OpenAI embeddings
+# Using config command (Recommended)
+codexa config set GROQ_API_KEY "gsk_your_key_here"
+
+# Or using environment variables
+export GROQ_API_KEY="gsk_your_key_here" # macOS/Linux
 ```
 
 ### Configuration Options
@@ -416,7 +431,7 @@ The LLM provider to use for generating answers.
 
 **Type:** `string`  
 **Type:** `string`  
-**Default:** `"llama-3.1-8b-instant"`
+**Default:** `"openai/gpt-oss-120b"`
 
 The model identifier to use.
 
@@ -532,7 +547,7 @@ Maximum file size in bytes. Files larger than this will be excluded from indexin
 **Example:**
 ```json
 {
-  "maxFileSize": 10485760  // 10MB
+  "maxFileSize": 10485760
 }
 ```
 
@@ -561,7 +576,7 @@ Whether to skip files exceeding `maxFileSize` during indexing. Set to `false` if
 ```json
 {
   "skipLargeFiles": true,
-  "maxFileSize": 10485760  // 10MB
+  "maxFileSize": 10485760
 }
 ```
 
@@ -572,7 +587,7 @@ Whether to skip files exceeding `maxFileSize` during indexing. Set to `false` if
 ```json
 {
   "modelProvider": "groq",
-  "model": "llama-3.1-8b-instant",
+  "model": "openai/gpt-oss-120b",
   "embeddingProvider": "local",
   "embeddingModel": "Xenova/all-MiniLM-L6-v2",
   "maxChunkSize": 300,
@@ -582,9 +597,9 @@ Whether to skip files exceeding `maxFileSize` during indexing. Set to `false` if
 }
 ```
 
-**Remember:** Set `GROQ_API_KEY` environment variable:
+**Remember:** Set `GROQ_API_KEY`:
 ```bash
-export GROQ_API_KEY="your-api-key"
+codexa config set GROQ_API_KEY "your-api-key"
 ```
 
 
@@ -594,7 +609,7 @@ export GROQ_API_KEY="your-api-key"
 ```json
 {
   "modelProvider": "groq",
-  "model": "llama-3.1-8b-instant",
+  "model": "openai/gpt-oss-120b",
   "maxChunkSize": 150,
   "chunkOverlap": 15,
   "topK": 6,
@@ -623,10 +638,13 @@ export GROQ_API_KEY="your-api-key"
 cd my-project
 codexa init
 
-# 2. Index your codebase
+# 2. Set Groq Api Key
+codexa config set GROQ_API_KEY <your-groq-key>
+
+# 3. Index your codebase
 codexa ingest
 
-# 3. Ask questions
+# 4. Ask questions
 codexa ask "What is the main purpose of this codebase?"
 codexa ask "How does the user authentication work?"
 codexa ask "Where is the API routing configured?"
@@ -748,14 +766,17 @@ When you run `codexa ask`:
 **Problem:** Using Groq provider but API key is missing.
 
 **Solutions:**
-1. Set the environment variable:
+1. Set the API key using the config command (Recommended):
+   ```bash
+   codexa config set GROQ_API_KEY "your-api-key"
+   ```
+2. Or set the environment variable:
    ```bash
-   export GROQ_API_KEY="your-api-key"
+   export GROQ_API_KEY="your-api-key" # macOS/Linux
    ```
-2. Or add it to your shell profile (`~/.bashrc`, `~/.zshrc`, etc.)
 3. Verify it's set:
    ```bash
-   echo $GROQ_API_KEY
+   codexa config get GROQ_API_KEY
    ```
 
 ### Ingestion is Very Slow

package/dist/cli.js

@@ -7,6 +7,8 @@ Object.defineProperty(exports, "__esModule", { value: true });
 const commander_1 = require("commander");
 const ora_1 = __importDefault(require("ora"));
 const chalk_1 = __importDefault(require("chalk"));
+const node_path_1 = __importDefault(require("node:path"));
+const fs_extra_1 = __importDefault(require("fs-extra"));
 const config_1 = require("./config");
 const ingest_1 = require("./ingest");
 const agent_1 = require("./agent");
@@ -14,29 +16,67 @@ const logger_1 = require("./utils/logger");
 const formatter_1 = require("./utils/formatter");
 const marked_1 = require("marked");
 const marked_terminal_1 = __importDefault(require("marked-terminal"));
+const gradient_string_1 = __importDefault(require("gradient-string"));
+const boxen_1 = __importDefault(require("boxen"));
 marked_1.marked.setOptions({
     renderer: new marked_terminal_1.default({
         tab: 2,
     }),
 });
+function showBanner() {
+    const asciiArt = `
+    ╔════════════════════════════════════════════════════════╗
+    ║                                                        ║
+    ║     ██████╗ ██████╗ ██████╗ ███████╗██╗  ██╗ █████╗    ║
+    ║    ██╔════╝██╔═══██╗██╔══██╗██╔════╝╚██╗██╔╝██╔══██╗   ║
+    ║    ██║     ██║   ██║██║  ██║█████╗   ╚███╔╝ ███████║   ║
+    ║    ██║     ██║   ██║██║  ██║██╔══╝   ██╔██╗ ██╔══██║   ║
+    ║    ╚██████╗╚██████╔╝██████╔╝███████╗██╔╝ ██╗██║  ██║   ║
+    ║     ╚═════╝ ╚═════╝ ╚═════╝ ╚══════╝╚═╝  ╚═╝╚═╝  ╚═╝   ║
+    ║                                                        ║
+    ╚════════════════════════════════════════════════════════╝
+  `;
+    try {
+        const gradientArt = (0, gradient_string_1.default)('cyan', 'blue', 'magenta')(asciiArt);
+        console.log(gradientArt);
+        const message = (0, boxen_1.default)(`${chalk_1.default.bold('🎉 Codexa installed successfully!')}\n\n` +
+            `${chalk_1.default.dim('Codexa is a CLI tool that helps you understand your codebase using AI.')}\n\n` +
+            `${chalk_1.default.bold('Quick Start:')}\n\n` +
+            `${chalk_1.default.dim('1.')} ${chalk_1.default.white('Navigate to your project directory')}\n` +
+            `${chalk_1.default.dim('2.')} ${chalk_1.default.white('Initialize Codexa:')} ${chalk_1.default.cyan('codexa init')}\n` +
+            `${chalk_1.default.dim('3.')} ${chalk_1.default.white('Set your GROQ API Key:')} ` +
+            `${getAPIKeyStep()}\n` +
+            `${chalk_1.default.dim('4.')} ${chalk_1.default.white('Index your codebase:')} ${chalk_1.default.cyan('codexa ingest')}\n` +
+            `${chalk_1.default.dim('5.')} ${chalk_1.default.white('Ask questions:')} ${chalk_1.default.cyan('codexa ask "your question"')}\n\n` +
+            `${chalk_1.default.dim('For help, run:')} ${chalk_1.default.cyan('codexa --help')}\n` +
+            `${chalk_1.default.dim('Or visit:')} ${chalk_1.default.cyan('https://github.com/sahitya-chandra/codexa')}`, {
+            title: '🚀 Welcome to Codexa',
+            borderColor: 'cyan',
+            padding: 1,
+            margin: 1,
+        });
+        console.log(message);
+    }
+    catch {
+        console.log('\n🎉 Codexa installed successfully!\n');
+        console.log('Quick Start:');
+        console.log('1. Navigate to your project directory');
+        console.log('2. Initialize Codexa: codexa init');
+        console.log('3. Set your GROQ API Key');
+        console.log('4. Index your codebase: codexa ingest');
+        console.log('5. Ask questions: codexa ask "your question"\n');
+    }
+}
+function getAPIKeyStep() {
+    return `${chalk_1.default.cyan('codexa config set GROQ_API_KEY <your-groq-key>')}`;
+}
 const program = new commander_1.Command();
 program
     .name('codexa')
     .description('Ask questions about any local repository from the command line.')
-    .version('1.2.0')
+    .version('1.2.2')
     .action(() => {
-    console.log('\n');
-    logger_1.log.box(`${chalk_1.default.bold('Welcome to Codexa!')}\n\n` +
-        `${chalk_1.default.dim('Codexa is a CLI tool that helps you understand your codebase using AI.')}\n\n` +
-        `${chalk_1.default.bold('Getting Started:')}\n\n` +
-        `${chalk_1.default.dim('1.')} ${chalk_1.default.white('Initialize Codexa in your project:')}\n` +
-        `   ${chalk_1.default.cyan('codexa init')}\n\n` +
-        `${chalk_1.default.dim('2.')} ${chalk_1.default.white('Index your codebase:')}\n` +
-        `   ${chalk_1.default.cyan('codexa ingest')}\n\n` +
-        `${chalk_1.default.dim('3.')} ${chalk_1.default.white('Ask questions:')}\n` +
-        `   ${chalk_1.default.cyan('codexa ask "your question"')}\n\n` +
-        `${chalk_1.default.dim('For more help, run:')} ${chalk_1.default.cyan('codexa --help')}`, '🚀 Codexa');
-    console.log('\n');
+    showBanner();
 });
 program
     .command('init')
@@ -49,8 +89,10 @@ program
     console.log('\n');
     logger_1.log.box(`${chalk_1.default.bold('Next Steps:')}\n\n` +
         `${chalk_1.default.dim('1.')} ${chalk_1.default.white('Review .codexarc.json')} - Update provider keys if needed\n` +
-        `${chalk_1.default.dim('2.')} ${chalk_1.default.white('Run:')} ${chalk_1.default.cyan('codexa ingest')} ${chalk_1.default.dim('- Start indexing your codebase')}\n` +
-        `${chalk_1.default.dim('3.')} ${chalk_1.default.white('Run:')} ${chalk_1.default.cyan('codexa ask "your question"')} ${chalk_1.default.dim('- Ask questions about your code')}`, '🚀 Setup Complete');
+        `${chalk_1.default.dim('2.')} ${chalk_1.default.white('Set your GROQ API Key:')} ` +
+        `${getAPIKeyStep()}\n` +
+        `${chalk_1.default.dim('3.')} ${chalk_1.default.white('Run:')} ${chalk_1.default.cyan('codexa ingest')} ${chalk_1.default.dim('- Start indexing your codebase')}\n` +
+        `${chalk_1.default.dim('4.')} ${chalk_1.default.white('Run:')} ${chalk_1.default.cyan('codexa ask "your question"')} ${chalk_1.default.dim('- Ask questions about your code')}`, '🚀 Setup Complete');
     console.log('\n');
 });
 program
@@ -114,6 +156,53 @@ program
         handleError(error);
     }
 });
+program
+    .command('config')
+    .description('Manage configuration values')
+    .argument('[action]', 'Action to perform (set, get, list)')
+    .argument('[key]', 'Configuration key')
+    .argument('[value]', 'Configuration value')
+    .action(async (action, key, value) => {
+    const cwd = process.cwd();
+    const configPath = node_path_1.default.join(cwd, config_1.CONFIG_FILENAME);
+    if (!(await fs_extra_1.default.pathExists(configPath))) {
+        logger_1.log.error(`Configuration file not found. Please run ${chalk_1.default.cyan('codexa init')} first.`);
+        return;
+    }
+    const config = await (0, config_1.loadConfig)(cwd);
+    if (action === 'set') {
+        if (!key || !value) {
+            logger_1.log.error('Usage: codexa config set <key> <value>');
+            return;
+        }
+        if (key === 'GROQ_API_KEY') {
+            config.groqApiKey = value;
+            await (0, config_1.saveConfig)(cwd, config);
+            logger_1.log.success(`Updated ${key}`);
+        }
+        else {
+            logger_1.log.error(`Unknown config key: ${key}`);
+        }
+    }
+    else if (action === 'get') {
+        if (!key) {
+            logger_1.log.error('Usage: codexa config get <key>');
+            return;
+        }
+        if (key === 'GROQ_API_KEY') {
+            console.log(config.groqApiKey || 'Not set');
+        }
+        else {
+            logger_1.log.error(`Unknown config key: ${key}`);
+        }
+    }
+    else if (action === 'list') {
+        console.log(config);
+    }
+    else {
+        logger_1.log.error('Usage: codexa config <set|get|list> [key] [value]');
+    }
+});
 program.parseAsync(process.argv).catch(handleError);
 function handleError(error) {
     if (error instanceof Error) {

package/dist/config/generator.js

@@ -352,7 +352,7 @@ function generateConfig(analysis, baseConfig) {
     // Start with base config or defaults
     const config = {
         modelProvider: 'groq',
-        model: 'llama-3.1-8b-instant',
+        model: 'openai/gpt-oss-120b',
         embeddingProvider: 'local',
         embeddingModel: 'Xenova/all-MiniLM-L6-v2',
         maxChunkSize: 800,

package/dist/config.js

@@ -3,6 +3,7 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
     return (mod && mod.__esModule) ? mod : { "default": mod };
 };
 Object.defineProperty(exports, "__esModule", { value: true });
+exports.CONFIG_FILENAME = void 0;
 exports.ensureConfig = ensureConfig;
 exports.loadConfig = loadConfig;
 exports.saveConfig = saveConfig;
@@ -13,10 +14,10 @@ const ora_1 = __importDefault(require("ora"));
 const detector_1 = require("./config/detector");
 const generator_1 = require("./config/generator");
 dotenv_1.default.config();
-const CONFIG_FILENAME = '.codexarc.json';
+exports.CONFIG_FILENAME = '.codexarc.json';
 const DEFAULT_CONFIG = {
     modelProvider: 'groq',
-    model: 'llama-3.1-8b-instant', // can also use llama-3.3-70b-versatile for better perf
+    model: 'openai/gpt-oss-120b', // can also use llama-3.3-70b-versatile for better perf
     embeddingProvider: 'local',
     embeddingModel: 'Xenova/all-MiniLM-L6-v2',
     // localModelUrl: 'http://localhost:11434',
@@ -66,7 +67,7 @@ async function generateDynamicConfig(cwd) {
     }
 }
 async function ensureConfig(cwd) {
-    const configPath = node_path_1.default.join(cwd, CONFIG_FILENAME);
+    const configPath = node_path_1.default.join(cwd, exports.CONFIG_FILENAME);
     if (!(await fs_extra_1.default.pathExists(configPath))) {
         const dynamicConfig = await generateDynamicConfig(cwd);
         await fs_extra_1.default.writeJson(configPath, dynamicConfig, { spaces: 2 });
@@ -74,7 +75,7 @@ async function ensureConfig(cwd) {
     return loadConfig(cwd);
 }
 async function loadConfig(cwd) {
-    const configPath = node_path_1.default.join(cwd, CONFIG_FILENAME);
+    const configPath = node_path_1.default.join(cwd, exports.CONFIG_FILENAME);
     let config;
     if (!(await fs_extra_1.default.pathExists(configPath))) {
         config = { ...DEFAULT_CONFIG };
@@ -88,7 +89,7 @@ async function loadConfig(cwd) {
     return hydratePaths(cwd, config);
 }
 async function saveConfig(cwd, config) {
-    const configPath = node_path_1.default.join(cwd, CONFIG_FILENAME);
+    const configPath = node_path_1.default.join(cwd, exports.CONFIG_FILENAME);
     const dehydrated = dehydratePaths(cwd, config);
     await fs_extra_1.default.writeJson(configPath, dehydrated, { spaces: 2 });
 }

package/dist/ingest.js

@@ -92,6 +92,10 @@ async function ingestRepository({ cwd, config, force = false, }) {
         spinnerFiles.fail('No matching files found.');
         return;
     }
+    if (allFiles.length > 200) {
+        spinnerFiles.stop();
+        throw new Error('Codebase is too large, cannot index it');
+    }
     // Filter files: exclude binaries and large files
     spinnerFiles.text = `Filtering files (found ${allFiles.length})...`;
     const { included: files, excluded } = await (0, file_filter_1.filterFiles)(allFiles, {
@@ -112,11 +116,22 @@ async function ingestRepository({ cwd, config, force = false, }) {
     const spinnerChunk = (0, ora_1.default)('Chunking files...').start();
     const chunks = [];
     for (const file of files) {
-        const ch = await (0, chunker_1.chunkFile)(file, config.maxChunkSize, config.chunkOverlap);
-        ch.forEach((c) => (c.filePath = node_path_1.default.relative(cwd, c.filePath)));
-        chunks.push(...ch);
+        try {
+            const ch = await (0, chunker_1.chunkFile)(file, config.maxChunkSize, config.chunkOverlap);
+            ch.forEach((c) => (c.filePath = node_path_1.default.relative(cwd, c.filePath)));
+            chunks.push(...ch);
+        }
+        catch (error) {
+            spinnerChunk.clear();
+            console.warn(`\nFailed to chunk file ${file}: ${error instanceof Error ? error.message : String(error)}`);
+            spinnerChunk.render();
+        }
         await tick();
     }
+    if (chunks.length > 20000) {
+        spinnerChunk.stop();
+        throw new Error('Chunk limit exceeded, unable to create embeddings!');
+    }
     spinnerChunk.succeed(`Chunked files (${chunks.length} chunks)`);
     const spinnerCompress = (0, ora_1.default)('Compressing chunks...').start();
     chunks.forEach((c) => (c.compressed = compressText(c.content)));
@@ -131,11 +146,22 @@ async function ingestRepository({ cwd, config, force = false, }) {
     const batchSize = 32;
     progress.start(chunks.length, 0);
     for (let i = 0; i < chunks.length; i += batchSize) {
-        const batch = chunks.slice(i, i + batchSize);
-        const texts = batch.map((c) => c.content);
-        const vectors = await embedder.embed(texts);
-        batch.forEach((c, idx) => (c.embedding = vectors[idx]));
-        progress.increment(batch.length);
+        try {
+            const batch = chunks.slice(i, i + batchSize);
+            const texts = batch.map((c) => c.content);
+            const vectors = await embedder.embed(texts);
+            batch.forEach((c, idx) => (c.embedding = vectors[idx]));
+            progress.increment(batch.length);
+        }
+        catch (error) {
+            // Log error but continue with next batch
+            // retry or DLQ this batch
+            const msg = error instanceof Error ? error.message : String(error);
+            // Temporarily stop progress bar to log error
+            progress.stop();
+            console.error(`\nFailed to embed batch starting at index ${i}: ${msg}`);
+            progress.start(chunks.length, i + batchSize);
+        }
         await tick();
     }
     progress.stop();

package/dist/models/index.js

@@ -59,8 +59,10 @@ class OllamaLLM {
                         options.onToken?.(token);
                     }
                 }
-                catch {
-                    // Should not happen with proper buffering, but ignore if it does
+                catch (e) {
+                    if (process.env.AGENT_DEBUG) {
+                        console.warn('Failed to parse token:', e);
+                    }
                 }
             }
         }
@@ -74,8 +76,11 @@ class OllamaLLM {
                     options.onToken?.(token);
                 }
             }
-            catch {
+            catch (e) {
                 // Ignore incomplete final chunk
+                if (process.env.AGENT_DEBUG) {
+                    console.warn('Failed to parse final chunk:', e);
+                }
             }
         }
         return full;
@@ -125,10 +130,11 @@ function createLLMClient(config) {
         if (process.env.AGENT_DEBUG) {
             console.error('Using Groq client:', config.model);
         }
-        if (!process.env.GROQ_API_KEY) {
-            throw new Error('GROQ_API_KEY is not set. Please set the GROQ_API_KEY environment variable to use Groq models.');
+        const apiKey = process.env.GROQ_API_KEY || config.groqApiKey;
+        if (!apiKey) {
+            throw new Error('GROQ_API_KEY is not set. Please set the GROQ_API_KEY environment variable or use `codexa config set GROQ_API_KEY <key>` to set it.');
         }
-        return new GroqLLM(config.model, process.env.GROQ_API_KEY);
+        return new GroqLLM(config.model, apiKey);
     }
     throw new Error('Only local provider supported for now.');
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "codexa",
-  "version": "1.2.0",
+  "version": "1.2.2",
   "description": "CLI agent that indexes local repos and answers questions with hosted or local LLMs.",
   "bin": {
     "codexa": "bin/codexa.js"
@@ -10,7 +10,6 @@
     "clean": "rimraf dist",
     "prepare": "npm run build",
     "prepublishOnly": "npm run clean && npm run build",
-    "postinstall": "node scripts/postinstall.js",
     "dev": "tsx src/cli.ts",
     "smoke": "ts-node --transpile-only scripts/smoke.ts",
     "lint": "eslint .",
@@ -19,10 +18,9 @@
   },
   "type": "commonjs",
   "keywords": [
-    "ai",
-    "cli",
-    "rag",
-    "codebase"
+    "ai", "cli", "rag", "codebase", "code-analysis",
+    "llm", "embeddings", "vector-search", "developer-tools",
+    "code-understanding", "semantic-search", "groq"
   ],
   "author": "",
   "license": "MIT",
@@ -37,7 +35,6 @@
   "files": [
     "bin",
     "dist",
-    "scripts",
     "README.md",
     "LICENSE"
   ],

package/scripts/postinstall.js

@@ -1,58 +0,0 @@
-#!/usr/bin/env node
-
-// Use dynamic import for ESM modules (chalk v5 is ESM-only)
-(async () => {
-  try {
-    const chalk = (await import('chalk')).default;
-    const boxen = (await import('boxen')).default;
-    const gradient = (await import('gradient-string')).default;
-
-    const asciiArt = `
-  ╔════════════════════════════════════════════════════════╗
-  ║                                                        ║
-  ║     ██████╗ ██████╗ ██████╗ ███████╗██╗  ██╗ █████╗    ║
-  ║    ██╔════╝██╔═══██╗██╔══██╗██╔════╝╚██╗██╔╝██╔══██╗   ║
-  ║    ██║     ██║   ██║██║  ██║█████╗   ╚███╔╝ ███████║   ║
-  ║    ██║     ██║   ██║██║  ██║██╔══╝   ██╔██╗ ██╔══██║   ║
-  ║    ╚██████╗╚██████╔╝██████╔╝███████╗██╔╝ ██╗██║  ██║   ║
-  ║     ╚═════╝ ╚═════╝ ╚═════╝ ╚══════╝╚═╝  ╚═╝╚═╝  ╚═╝   ║
-  ║                                                        ║
-  ╚════════════════════════════════════════════════════════╝
-`;
-
-    const gradientArt = gradient('cyan', 'blue', 'magenta')(asciiArt);
-
-    console.log('\n');
-    console.log(gradientArt);
-    console.log('\n');
-
-    const message = boxen(
-      `${chalk.bold('🎉 Codexa installed successfully!')}\n\n` +
-        `${chalk.dim('Codexa is a CLI tool that helps you understand your codebase using AI.')}\n\n` +
-        `${chalk.bold('Quick Start:')}\n\n` +
-        `${chalk.dim('1.')} ${chalk.white('Navigate to your project directory')}\n` +
-        `${chalk.dim('2.')} ${chalk.white('Initialize Codexa:')} ${chalk.cyan('codexa init')}\n` +
-        `${chalk.dim('3.')} ${chalk.white('Index your codebase:')} ${chalk.cyan('codexa ingest')}\n` +
-        `${chalk.dim('4.')} ${chalk.white('Ask questions:')} ${chalk.cyan('codexa ask "your question"')}\n\n` +
-        `${chalk.dim('For help, run:')} ${chalk.cyan('codexa --help')}\n` +
-        `${chalk.dim('Or visit:')} ${chalk.cyan('https://github.com/sahitya-chandra/codexa')}`,
-      {
-        title: '🚀 Welcome to Codexa',
-        borderColor: 'cyan',
-        padding: 1,
-        margin: 1,
-      },
-    );
-
-    console.log(message);
-    console.log('\n');
-  } catch {
-    // Fallback if modules aren't available (shouldn't happen, but just in case)
-    console.log('\n🎉 Codexa installed successfully!\n');
-    console.log('Quick Start:');
-    console.log('1. Navigate to your project directory');
-    console.log('2. Initialize Codexa: codexa init');
-    console.log('3. Index your codebase: codexa ingest');
-    console.log('4. Ask questions: codexa ask "your question"\n');
-  }
-})();

package/scripts/smoke.js

@@ -1,26 +0,0 @@
-'use strict';
-var __importDefault =
-  (this && this.__importDefault) ||
-  function (mod) {
-    return mod && mod.__esModule ? mod : { default: mod };
-  };
-Object.defineProperty(exports, '__esModule', { value: true });
-const node_path_1 = __importDefault(require('node:path'));
-const config_1 = require('../src/config');
-const ingest_1 = require('../src/ingest');
-const retriever_1 = require('../src/retriever');
-async function main() {
-  const sampleRoot = node_path_1.default.resolve(__dirname, '../examples/sample');
-  await (0, config_1.ensureConfig)(sampleRoot);
-  const config = await (0, config_1.loadConfig)(sampleRoot);
-  await (0, ingest_1.ingestRepository)({ cwd: sampleRoot, config, force: true });
-  const results = await (0, retriever_1.retrieveContext)('How do we greet someone?', config);
-  if (results.length === 0) {
-    throw new Error('Smoke test failed: no retrieval results.');
-  }
-  console.log('Top chunk:', results[0].filePath, `${results[0].startLine}-${results[0].endLine}`);
-}
-main().catch((error) => {
-  console.error(error);
-  process.exit(1);
-});

package/scripts/smoke.ts

@@ -1,21 +0,0 @@
-import path from 'node:path';
-import { ensureConfig, loadConfig } from '../src/config';
-import { ingestRepository } from '../src/ingest';
-import { retrieveContext } from '../src/retriever';
-
-async function main(): Promise<void> {
-  const sampleRoot = path.resolve(__dirname, '../examples/sample');
-  await ensureConfig(sampleRoot);
-  const config = await loadConfig(sampleRoot);
-  await ingestRepository({ cwd: sampleRoot, config, force: true });
-  const results = await retrieveContext('How do we greet someone?', config);
-  if (results.length === 0) {
-    throw new Error('Smoke test failed: no retrieval results.');
-  }
-  console.log('Top chunk:', results[0].filePath, `${results[0].startLine}-${results[0].endLine}`);
-}
-
-main().catch((error) => {
-  console.error(error);
-  process.exit(1);
-});