grok-dev 1.1.0 → 1.1.1

package/.claude/worktrees/vigilant-johnson/biome.json

@@ -0,0 +1,51 @@
+{
+  "$schema": "https://biomejs.dev/schemas/2.4.8/schema.json",
+  "vcs": {
+    "enabled": true,
+    "clientKind": "git",
+    "useIgnoreFile": true
+  },
+  "files": {
+    "includes": ["**", "!!**/dist"]
+  },
+  "formatter": {
+    "enabled": true,
+    "indentStyle": "space",
+    "indentWidth": 2,
+    "lineWidth": 120
+  },
+  "linter": {
+    "enabled": true,
+    "rules": {
+      "recommended": true,
+      "suspicious": {
+        "noExplicitAny": "error",
+        "noArrayIndexKey": "warn"
+      },
+      "correctness": {
+        "noUnusedFunctionParameters": "warn"
+      },
+      "complexity": {
+        "noForEach": "off"
+      },
+      "style": {
+        "noNonNullAssertion": "off",
+        "useNodejsImportProtocol": "off"
+      }
+    }
+  },
+  "javascript": {
+    "formatter": {
+      "quoteStyle": "double",
+      "trailingCommas": "all"
+    }
+  },
+  "assist": {
+    "enabled": true,
+    "actions": {
+      "source": {
+        "organizeImports": "on"
+      }
+    }
+  }
+}

package/.claude/worktrees/vigilant-johnson/package.json

@@ -0,0 +1,74 @@
+{
+  "name": "grok-dev",
+  "version": "1.1.0",
+  "description": "An open-source AI coding agent powered by Grok, built with Bun and OpenTUI.",
+  "type": "module",
+  "main": "dist/index.js",
+  "exports": {
+    ".": {
+      "import": "./dist/index.js",
+      "types": "./dist/index.d.ts"
+    }
+  },
+  "bin": {
+    "grok": "dist/index.js"
+  },
+  "scripts": {
+    "dev": "bun run src/index.ts",
+    "build": "tsc",
+    "start": "bun run dist/index.js",
+    "typecheck": "tsc --noEmit",
+    "test": "bunx vitest run",
+    "test:watch": "bunx vitest",
+    "lint": "biome check src/",
+    "format": "biome format src/",
+    "format:fix": "biome format --write src/",
+    "lint:fix": "biome check --fix src/",
+    "pre-commit": "lint-staged",
+    "prepare": "husky"
+  },
+  "lint-staged": {
+    "*.{ts,tsx,js,mjs,cjs,json}": "biome check --write --no-errors-on-unmatched"
+  },
+  "keywords": [
+    "cli",
+    "agent",
+    "grok",
+    "ai",
+    "coding",
+    "terminal",
+    "opentui"
+  ],
+  "author": "Vibe Kit",
+  "license": "MIT",
+  "dependencies": {
+    "@ai-sdk/mcp": "^1.0.25",
+    "@ai-sdk/xai": "^3.0.67",
+    "@modelcontextprotocol/sdk": "^1.27.1",
+    "@opentui/core": "^0.1.88",
+    "@opentui/react": "^0.1.88",
+    "ai": "^6.0.116",
+    "commander": "^12.1.0",
+    "diff": "^8.0.3",
+    "dotenv": "^16.6.1",
+    "grammy": "^1.41.1",
+    "react": "^19.2.4",
+    "semver": "^7.7.4",
+    "zod": "^4.3.6"
+  },
+  "devDependencies": {
+    "@biomejs/biome": "^2.4.8",
+    "@types/diff": "^8.0.0",
+    "@types/node": "^22.19.15",
+    "@types/react": "^19.2.14",
+    "@types/semver": "^7.7.1",
+    "husky": "^9.1.7",
+    "lint-staged": "^16.4.0",
+    "typescript": "^5.9.3",
+    "vitest": "^4.1.0"
+  },
+  "engines": {
+    "node": ">=18.0.0"
+  },
+  "preferGlobal": true
+}

package/.claude/worktrees/vigilant-johnson/vitest.config.ts

@@ -0,0 +1,7 @@
+import { defineConfig } from "vitest/config";
+
+export default defineConfig({
+  test: {
+    exclude: ["dist/**", "node_modules/**"],
+  },
+});

package/README.md

@@ -60,8 +60,14 @@ grok --prompt "run the test suite and summarize failures"
 grok -p "show me package.json" --directory /path/to/project
 grok --prompt "refactor X" --max-tool-rounds 30
 grok --prompt "summarize the repo state" --format json
+grok --prompt "review the repo overnight" --batch-api
+grok --verify
 ```
 
+`--batch-api` uses xAI's Batch API for lower-cost unattended runs. It is a good
+fit for scripts, CI, schedules, and other non-interactive workflows where a
+delayed result is fine.
+
 **Continue a saved session:**
 
 ```bash
@@ -135,6 +141,7 @@ You keep using a text model for the session, and Grok saves generated media unde
 | **X + web search** | **`search_x`** and **`search_web`** tools—live posts and docs without pretending the internet stopped in 2023. |
 | **Media generation** | Built-in **`generate_image`** and **`generate_video`** tools for text-to-image, image editing, text-to-video, and image-to-video flows. Generated files are saved locally so you can reuse them after the xAI URLs expire. |
 | **Sub-agents (default behavior)** | Foreground **`task`** delegation (e.g. explore vs general) plus background **`delegate`** for read-only deep dives—parallelize like you mean it. |
+| **Verify** | **`/verify`** or **`--verify`** — inspects your app, builds, tests, boots it, and runs browser smoke checks in a sandboxed environment. Screenshots and video included. |
 | **Custom sub-agents** | Define named agents with **`subAgents`** in **`~/.grok/user-settings.json`** and manage them from the TUI with **`/agents`**. |
 | **Remote control** | Pair **Telegram** from the TUI (`/remote-control` → Telegram): DM your bot, **`/pair`**, approve the code in-terminal. Keep the CLI running while you ping it from your phone. |
 | **No “mystery meat” UI** | OpenTUI React terminal UI—fast, keyboard-driven, not whatever glitchy thing you’re thinking of. |
@@ -146,7 +153,7 @@ You keep using a text model for the session, and Grok saves generated media unde
 
 ### Coming soon
 
-**Autonomous agent testing** (think: sandboxed machine, recorded runs, Replit-style “prove it works”—the kind of thing that makes flaky human QA nervous). Not shipped yet; when it lands, we’ll be insufferable about it.
+**Deeper autonomous agent testing** — persistent sandbox sessions, richer browser workflows, and stronger "prove it works" evidence.
 
 ---
 
@@ -190,7 +197,7 @@ Optional **`subAgents`** — custom foreground sub-agents. Each entry needs **`n
 }
 ```
 
-Names cannot be `general` or `explore` because those are reserved for the built-in sub-agents.
+Names cannot be `general`, `explore`, `vision`, or `verify` because those are reserved for the built-in sub-agents.
 
 Optional: **`GROK_BASE_URL`** (default `https://api.x.ai/v1`), **`GROK_MODEL`**, **`GROK_MAX_TOKENS`**.
 
@@ -300,6 +307,17 @@ When sandbox mode is active you can configure:
 
 All settings are saved in `~/.grok/user-settings.json` (user) and `.grok/settings.json` (project).
 
+### Verify
+
+Run **`/verify`** in the TUI or **`--verify`** on the CLI to verify your app locally:
+
+```bash
+grok --verify
+grok -d /path/to/your/app --verify
+```
+
+The agent inspects your project, figures out how to build and run it, spins up a sandbox, and produces a verification report with screenshots and video evidence. Works with any app type.
+
 ---
 
 ## Development

package/dist/agent/agent.d.ts

@@ -1,17 +1,19 @@
 import { type ScheduleDaemonStatus, type StoredSchedule } from "../tools/schedule";
-import type { AgentMode, ChatEntry, SessionInfo, SessionSnapshot, StreamChunk, SubagentStatus, TaskRequest, ToolCall, ToolResult } from "../types/index";
+import type { AgentMode, ChatEntry, SessionInfo, SessionSnapshot, StreamChunk, SubagentStatus, TaskRequest, ToolCall, ToolResult, VerifyRecipe } from "../types/index";
 import { type SandboxMode, type SandboxSettings } from "../utils/settings";
 interface AgentOptions {
     persistSession?: boolean;
     session?: string;
     sandboxMode?: SandboxMode;
     sandboxSettings?: SandboxSettings;
+    batchApi?: boolean;
 }
 type ProcessMessageFinishReason = "stop" | "length" | "content-filter" | "tool-calls" | "error" | "other";
 export interface ProcessMessageUsage {
     inputTokens?: number;
     outputTokens?: number;
     totalTokens?: number;
+    costUsdTicks?: number;
 }
 export interface ProcessMessageStepStart {
     stepNumber: number;
@@ -63,6 +65,7 @@ export declare class Agent {
     private planContext;
     private subagentStatusListeners;
     private sendTelegramFile;
+    private batchApi;
     constructor(apiKey: string | undefined, baseURL?: string, model?: string, maxToolRounds?: number, options?: AgentOptions);
     getModel(): string;
     setModel(model: string): void;
@@ -101,6 +104,9 @@ export declare class Agent {
     private discardAbortedTurn;
     private recordUsage;
     consumeBackgroundNotifications(): Promise<string[]>;
+    private getBatchClientOptions;
+    private executeBatchToolCall;
+    private runTaskRequestBatch;
     runTaskRequest(request: TaskRequest, onActivity?: (detail: string) => void, abortSignal?: AbortSignal): Promise<ToolResult>;
     private runTask;
     private runDelegation;
@@ -108,8 +114,10 @@ export declare class Agent {
     private listDelegations;
     private getCompactionSettings;
     private compactForContext;
+    private processMessageBatchTurn;
     private appendCompletedTurn;
     processMessage(userMessage: string, observer?: ProcessMessageObserver): AsyncGenerator<StreamChunk, void, unknown>;
     private requireProvider;
+    detectVerifyRecipe(settings?: SandboxSettings, abortSignal?: AbortSignal): Promise<VerifyRecipe | null>;
 }
 export {};