codemaxxing 0.4.16 → 1.0.0

package/README.md

@@ -191,10 +191,35 @@ Conversations auto-save to SQLite. Pick up where you left off:
 - `/session delete` — remove a session
 - `/resume` — interactive session picker
 
+### 🔌 MCP Support (Model Context Protocol)
+Connect to external tools via the industry-standard MCP protocol. Databases, GitHub, Slack, browsers — anything with an MCP server.
+- Compatible with `.cursor/mcp.json` and `opencode.json` configs
+- `/mcp` — show connected servers
+- `/mcp add github npx -y @modelcontextprotocol/server-github` — add a server
+- `/mcp tools` — list all available MCP tools
+
+### 🖥️ Zero-Setup Local LLM
+First time with no LLM? Codemaxxing walks you through it:
+1. Detects your hardware (CPU, RAM, GPU)
+2. Recommends coding models that fit your machine
+3. Installs Ollama automatically
+4. Downloads the model with a progress bar
+5. Connects and drops you into coding mode
+
+No googling, no config files, no decisions. Just run `codemaxxing`.
+
+### 🦙 Ollama Management
+Full Ollama control from inside codemaxxing:
+- `/ollama` — status, installed models, GPU usage
+- `/ollama pull` — interactive model picker + download
+- `/ollama delete` — pick and remove models
+- `/ollama start` / `/ollama stop` — server management
+- Exit warning when Ollama is using GPU memory
+
 ### 🔄 Multi-Provider
-Switch models mid-session without restarting:
-- `/model gpt-4o` — switch to a different model
-- `/models` — list available models from your provider
+Switch models mid-session with an interactive picker:
+- `/model` — browse and switch models
+- `/model gpt-4o` — switch directly by name
 - Native Anthropic API support (not just OpenAI-compatible)
 
 ### 🎨 14 Themes
@@ -216,11 +241,15 @@ Type `/` for autocomplete suggestions. Arrow keys to navigate, Tab or Enter to s
 | `/help` | Show all commands |
 | `/connect` | Retry LLM connection |
 | `/login` | Interactive auth setup |
+| `/model` | Browse & switch models (picker) |
 | `/architect` | Toggle architect mode / set model |
 | `/skills` | Browse, install, manage skills |
 | `/lint on/off` | Toggle auto-linting |
-| `/model <name>` | Switch model mid-session |
-| `/models` | List available models |
+| `/mcp` | MCP server status & tools |
+| `/ollama` | Ollama status, models & GPU |
+| `/ollama pull` | Download a model (picker) |
+| `/ollama delete` | Remove a model (picker) |
+| `/ollama start/stop` | Server management |
 | `/theme` | Switch color theme |
 | `/map` | Show repository map |
 | `/sessions` | List past sessions |
@@ -295,14 +324,16 @@ Settings are stored in `~/.codemaxxing/settings.json`:
 
 ## Tools
 
-Codemaxxing gives the model these tools:
+Built-in tools:
 
 - **read_file** — Read file contents (safe)
-- **write_file** — Write/create files (requires approval)
+- **write_file** — Write/create files (requires approval, shows diff)
 - **list_files** — List directory contents (safe)
 - **search_files** — Search for patterns across files (safe)
 - **run_command** — Execute shell commands (requires approval)
 
+Plus any tools from connected MCP servers (databases, APIs, GitHub, etc.)
+
 ## Project Context
 
 Drop a `CODEMAXXING.md` file in your project root to give the model extra context about your codebase, conventions, or instructions. It's automatically included in the system prompt.
@@ -311,8 +342,10 @@ Drop a `CODEMAXXING.md` file in your project root to give the model extra contex
 
 - **Runtime:** Node.js + TypeScript
 - **TUI:** [Ink](https://github.com/vadimdemedes/ink) (React for the terminal)
-- **LLM SDK:** [OpenAI SDK](https://github.com/openai/openai-node) (works with any compatible API)
+- **LLM SDKs:** [OpenAI SDK](https://github.com/openai/openai-node) + [Anthropic SDK](https://github.com/anthropics/anthropic-sdk-typescript)
+- **MCP:** [@modelcontextprotocol/sdk](https://github.com/modelcontextprotocol/typescript-sdk)
 - **Sessions:** [better-sqlite3](https://github.com/WiseLibs/better-sqlite3)
+- **Local LLM:** Ollama integration (auto-install, pull, manage)
 - **Zero cloud dependencies** — everything runs locally
 
 ## Inspired By

package/dist/config.d.ts

@@ -41,6 +41,17 @@ export declare function getConfigPath(): string;
 /**
  * Auto-detect local LLM servers
  */
+export type DetectionResult = {
+    status: "connected";
+    provider: ProviderConfig;
+} | {
+    status: "no-models";
+    serverName: string;
+    baseUrl: string;
+} | {
+    status: "no-server";
+};
+export declare function detectLocalProviderDetailed(): Promise<DetectionResult>;
 export declare function detectLocalProvider(): Promise<ProviderConfig | null>;
 /**
  * List available models from a provider endpoint

package/dist/config.js

@@ -149,9 +149,49 @@ export function applyOverrides(config, args) {
 export function getConfigPath() {
     return CONFIG_FILE;
 }
-/**
- * Auto-detect local LLM servers
- */
+export async function detectLocalProviderDetailed() {
+    const endpoints = [
+        { name: "LM Studio", url: "http://localhost:1234/v1" },
+        { name: "Ollama", url: "http://localhost:11434/v1" },
+        { name: "vLLM", url: "http://localhost:8000/v1" },
+    ];
+    let serverFound = null;
+    for (const endpoint of endpoints) {
+        try {
+            const controller = new AbortController();
+            const timeout = setTimeout(() => controller.abort(), 2000);
+            const res = await fetch(`${endpoint.url}/models`, {
+                signal: controller.signal,
+            });
+            clearTimeout(timeout);
+            if (res.ok) {
+                const data = (await res.json());
+                const models = data.data ?? [];
+                if (models.length === 0) {
+                    // Server is up but no models — remember it but keep looking
+                    if (!serverFound)
+                        serverFound = endpoint;
+                    continue;
+                }
+                return {
+                    status: "connected",
+                    provider: {
+                        baseUrl: endpoint.url,
+                        apiKey: "not-needed",
+                        model: models[0].id,
+                    },
+                };
+            }
+        }
+        catch {
+            // Server not running, try next
+        }
+    }
+    if (serverFound) {
+        return { status: "no-models", serverName: serverFound.name, baseUrl: serverFound.url };
+    }
+    return { status: "no-server" };
+}
 export async function detectLocalProvider() {
     const endpoints = [
         { name: "LM Studio", url: "http://localhost:1234/v1" },
@@ -169,7 +209,11 @@ export async function detectLocalProvider() {
             if (res.ok) {
                 const data = (await res.json());
                 const models = data.data ?? [];
-                const model = models[0]?.id ?? "auto";
+                if (models.length === 0) {
+                    // Server is up but no models available — don't fake a connection
+                    continue;
+                }
+                const model = models[0].id;
                 return {
                     baseUrl: endpoint.url,
                     apiKey: "not-needed",

package/dist/index.js

@@ -5,7 +5,7 @@ import { render, Box, Text, useInput, useApp, useStdout } from "ink";
 import { EventEmitter } from "events";
 import TextInput from "ink-text-input";
 import { CodingAgent } from "./agent.js";
-import { loadConfig, saveConfig, detectLocalProvider, parseCLIArgs, applyOverrides, listModels } from "./config.js";
+import { loadConfig, saveConfig, detectLocalProvider, detectLocalProviderDetailed, parseCLIArgs, applyOverrides, listModels } from "./config.js";
 import { listSessions, getSession, loadMessages, deleteSession } from "./utils/sessions.js";
 import { execSync } from "child_process";
 import { isGitRepo, getBranch, getStatus, getDiff, undoLastCommit } from "./utils/git.js";
@@ -79,11 +79,37 @@ const SLASH_COMMANDS = [
 ];
 const SPINNER_FRAMES = ["⣾", "⣽", "⣻", "⢿", "⡿", "⣟", "⣯", "⣷"];
 const SPINNER_MESSAGES = [
+    // OG
     "Locking in...", "Cooking...", "Maxxing...", "In the zone...",
     "Yapping...", "Frame mogging...", "Jester gooning...", "Gooning...",
     "Doing back flips...", "Jester maxxing...", "Getting baked...",
     "Blasting tren...", "Pumping...", "Wondering if I should actually do this...",
     "Hacking the main frame...", "Codemaxxing...", "Vibe coding...", "Running a marathon...",
+    // Gym/Looksmaxxing
+    "Mewing aggressively...", "Looksmaxxing your codebase...", "Hitting a PR on this function...",
+    "Eating 4000 calories of code...", "Creatine loading...", "On my bulk arc...",
+    "Warming up the deadlift...",
+    // Brainrot/Skibidi
+    "Going full skibidi...", "Sigma grinding...", "Rizzing up the compiler...",
+    "No cap processing...", "Main character coding...", "It's giving implementation...",
+    "This code is bussin fr fr...", "Aura check in progress...", "Erm what the sigma...",
+    // Deranged/Unhinged
+    "Ascending to a higher plane...", "Achieving final form...", "Third eye compiling...",
+    "Astral projecting through your repo...", "Becoming one with the codebase...",
+    "Having a spiritual awakening...", "Entering the shadow realm...", "Going goblin mode...",
+    "Deleting System32... jk...", "Sacrificing tokens to the GPU gods...",
+    "Summoning the machine spirit...",
+    // Self-aware/Meta
+    "Pretending to think really hard...", "Staring at your code judgmentally...",
+    "Rethinking my career choices...", "Having an existential crisis...",
+    "Hoping this actually works...", "Praying to the stack overflow gods...",
+    "Copying from the internet with dignity...",
+    // Pure Chaos
+    "Doing hot yoga in the terminal...", "Microdosing your dependencies...",
+    "Running on 3 hours of sleep...", "Speedrunning your deadline...",
+    "Built different rn...", "That's crazy let me cook...",
+    "Absolutely feral right now...", "Ong no cap fr fr...",
+    "Living rent free in your RAM...", "Ate and left no crumbs...",
 ];
 // ── Neon Spinner ──
 function NeonSpinner({ message, colors }) {
@@ -220,15 +246,21 @@ function App() {
         if (provider.model === "auto" || (provider.baseUrl === "http://localhost:1234/v1" && !cliArgs.baseUrl)) {
             info.push("Detecting local LLM server...");
             setConnectionInfo([...info]);
-            const detected = await detectLocalProvider();
-            if (detected) {
+            const detection = await detectLocalProviderDetailed();
+            if (detection.status === "connected") {
                 // Keep CLI model override if specified
                 if (cliArgs.model)
-                    detected.model = cliArgs.model;
-                provider = detected;
+                    detection.provider.model = cliArgs.model;
+                provider = detection.provider;
                 info.push(`✔ Connected to ${provider.baseUrl} → ${provider.model}`);
                 setConnectionInfo([...info]);
             }
+            else if (detection.status === "no-models") {
+                info.push(`⚠ ${detection.serverName} is running but has no models. Use /ollama pull to download one.`);
+                setConnectionInfo([...info]);
+                setReady(true);
+                return;
+            }
             else {
                 info.push("✗ No local LLM server found.");
                 setConnectionInfo([...info]);
@@ -903,8 +935,9 @@ function App() {
             }
             return;
         }
-        if (trimmed === "/model" || trimmed === "/models") {
+        if (trimmed === "/model") {
             // Show picker of available models
+            addMsg("info", "Fetching available models...");
             try {
                 const ollamaModels = await listInstalledModelsDetailed();
                 if (ollamaModels.length > 0) {
@@ -913,18 +946,25 @@ function App() {
                     return;
                 }
             }
-            catch { }
+            catch (err) {
+                // Ollama not available or failed, try provider
+            }
             // Fallback: try provider's model list
-            try {
-                const providerModels = await listModels(providerRef.current?.baseUrl || "", providerRef.current?.apiKey || "");
-                if (providerModels.length > 0) {
-                    setModelPicker(providerModels.map((m) => m.id || m));
-                    setModelPickerIndex(0);
-                    return;
+            if (providerRef.current?.baseUrl && providerRef.current.baseUrl !== "auto") {
+                try {
+                    const providerModels = await listModels(providerRef.current.baseUrl, providerRef.current.apiKey || "");
+                    if (providerModels.length > 0) {
+                        setModelPicker(providerModels);
+                        setModelPickerIndex(0);
+                        return;
+                    }
+                }
+                catch (err) {
+                    // Provider fetch failed
                 }
             }
-            catch { }
-            addMsg("info", `Current model: ${modelName}\n  Usage: /model <model-name>`);
+            // No models found anywhere
+            addMsg("error", "No models available. Download one with /ollama pull or configure a provider.");
             return;
         }
         if (trimmed.startsWith("/model ")) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "codemaxxing",
-  "version": "0.4.16",
+  "version": "1.0.0",
   "description": "Open-source terminal coding agent. Connect any LLM. Max your code.",
   "main": "dist/index.js",
   "bin": {

package/src/config.ts

@@ -193,6 +193,57 @@ export function getConfigPath(): string {
 /**
  * Auto-detect local LLM servers
  */
+export type DetectionResult =
+  | { status: "connected"; provider: ProviderConfig }
+  | { status: "no-models"; serverName: string; baseUrl: string }
+  | { status: "no-server" };
+
+export async function detectLocalProviderDetailed(): Promise<DetectionResult> {
+  const endpoints = [
+    { name: "LM Studio", url: "http://localhost:1234/v1" },
+    { name: "Ollama", url: "http://localhost:11434/v1" },
+    { name: "vLLM", url: "http://localhost:8000/v1" },
+  ];
+
+  let serverFound: { name: string; url: string } | null = null;
+
+  for (const endpoint of endpoints) {
+    try {
+      const controller = new AbortController();
+      const timeout = setTimeout(() => controller.abort(), 2000);
+      const res = await fetch(`${endpoint.url}/models`, {
+        signal: controller.signal,
+      });
+      clearTimeout(timeout);
+
+      if (res.ok) {
+        const data = (await res.json()) as { data?: Array<{ id: string }> };
+        const models = data.data ?? [];
+        if (models.length === 0) {
+          // Server is up but no models — remember it but keep looking
+          if (!serverFound) serverFound = endpoint;
+          continue;
+        }
+        return {
+          status: "connected",
+          provider: {
+            baseUrl: endpoint.url,
+            apiKey: "not-needed",
+            model: models[0]!.id,
+          },
+        };
+      }
+    } catch {
+      // Server not running, try next
+    }
+  }
+
+  if (serverFound) {
+    return { status: "no-models", serverName: serverFound.name, baseUrl: serverFound.url };
+  }
+  return { status: "no-server" };
+}
+
 export async function detectLocalProvider(): Promise<ProviderConfig | null> {
   const endpoints = [
     { name: "LM Studio", url: "http://localhost:1234/v1" },
@@ -212,7 +263,11 @@ export async function detectLocalProvider(): Promise<ProviderConfig | null> {
       if (res.ok) {
         const data = (await res.json()) as { data?: Array<{ id: string }> };
         const models = data.data ?? [];
-        const model = models[0]?.id ?? "auto";
+        if (models.length === 0) {
+          // Server is up but no models available — don't fake a connection
+          continue;
+        }
+        const model = models[0]!.id;
         return {
           baseUrl: endpoint.url,
           apiKey: "not-needed",

package/src/index.tsx

@@ -5,7 +5,7 @@ import { render, Box, Text, useInput, useApp, useStdout } from "ink";
 import { EventEmitter } from "events";
 import TextInput from "ink-text-input";
 import { CodingAgent } from "./agent.js";
-import { loadConfig, saveConfig, detectLocalProvider, parseCLIArgs, applyOverrides, listModels } from "./config.js";
+import { loadConfig, saveConfig, detectLocalProvider, detectLocalProviderDetailed, parseCLIArgs, applyOverrides, listModels } from "./config.js";
 import { listSessions, getSession, loadMessages, deleteSession } from "./utils/sessions.js";
 import { execSync } from "child_process";
 import { isGitRepo, getBranch, getStatus, getDiff, undoLastCommit } from "./utils/git.js";
@@ -81,11 +81,37 @@ const SLASH_COMMANDS = [
 const SPINNER_FRAMES = ["⣾", "⣽", "⣻", "⢿", "⡿", "⣟", "⣯", "⣷"];
 
 const SPINNER_MESSAGES = [
+  // OG
   "Locking in...", "Cooking...", "Maxxing...", "In the zone...",
   "Yapping...", "Frame mogging...", "Jester gooning...", "Gooning...",
   "Doing back flips...", "Jester maxxing...", "Getting baked...",
   "Blasting tren...", "Pumping...", "Wondering if I should actually do this...",
   "Hacking the main frame...", "Codemaxxing...", "Vibe coding...", "Running a marathon...",
+  // Gym/Looksmaxxing
+  "Mewing aggressively...", "Looksmaxxing your codebase...", "Hitting a PR on this function...",
+  "Eating 4000 calories of code...", "Creatine loading...", "On my bulk arc...",
+  "Warming up the deadlift...",
+  // Brainrot/Skibidi
+  "Going full skibidi...", "Sigma grinding...", "Rizzing up the compiler...",
+  "No cap processing...", "Main character coding...", "It's giving implementation...",
+  "This code is bussin fr fr...", "Aura check in progress...", "Erm what the sigma...",
+  // Deranged/Unhinged
+  "Ascending to a higher plane...", "Achieving final form...", "Third eye compiling...",
+  "Astral projecting through your repo...", "Becoming one with the codebase...",
+  "Having a spiritual awakening...", "Entering the shadow realm...", "Going goblin mode...",
+  "Deleting System32... jk...", "Sacrificing tokens to the GPU gods...",
+  "Summoning the machine spirit...",
+  // Self-aware/Meta
+  "Pretending to think really hard...", "Staring at your code judgmentally...",
+  "Rethinking my career choices...", "Having an existential crisis...",
+  "Hoping this actually works...", "Praying to the stack overflow gods...",
+  "Copying from the internet with dignity...",
+  // Pure Chaos
+  "Doing hot yoga in the terminal...", "Microdosing your dependencies...",
+  "Running on 3 hours of sleep...", "Speedrunning your deadline...",
+  "Built different rn...", "That's crazy let me cook...",
+  "Absolutely feral right now...", "Ong no cap fr fr...",
+  "Living rent free in your RAM...", "Ate and left no crumbs...",
 ];
 
 // ── Neon Spinner ──
@@ -264,13 +290,18 @@ function App() {
     if (provider.model === "auto" || (provider.baseUrl === "http://localhost:1234/v1" && !cliArgs.baseUrl)) {
       info.push("Detecting local LLM server...");
       setConnectionInfo([...info]);
-      const detected = await detectLocalProvider();
-      if (detected) {
+      const detection = await detectLocalProviderDetailed();
+      if (detection.status === "connected") {
         // Keep CLI model override if specified
-        if (cliArgs.model) detected.model = cliArgs.model;
-        provider = detected;
+        if (cliArgs.model) detection.provider.model = cliArgs.model;
+        provider = detection.provider;
         info.push(`✔ Connected to ${provider.baseUrl} → ${provider.model}`);
         setConnectionInfo([...info]);
+      } else if (detection.status === "no-models") {
+        info.push(`⚠ ${detection.serverName} is running but has no models. Use /ollama pull to download one.`);
+        setConnectionInfo([...info]);
+        setReady(true);
+        return;
       } else {
         info.push("✗ No local LLM server found.");
         setConnectionInfo([...info]);
@@ -941,8 +972,9 @@ function App() {
       }
       return;
     }
-    if (trimmed === "/model" || trimmed === "/models") {
+    if (trimmed === "/model") {
       // Show picker of available models
+      addMsg("info", "Fetching available models...");
       try {
         const ollamaModels = await listInstalledModelsDetailed();
         if (ollamaModels.length > 0) {
@@ -950,17 +982,26 @@ function App() {
           setModelPickerIndex(0);
           return;
         }
-      } catch {}
+      } catch (err) {
+        // Ollama not available or failed, try provider
+      }
+      
       // Fallback: try provider's model list
-      try {
-        const providerModels = await listModels(providerRef.current?.baseUrl || "", providerRef.current?.apiKey || "");
-        if (providerModels.length > 0) {
-          setModelPicker(providerModels.map((m: any) => m.id || m));
-          setModelPickerIndex(0);
-          return;
+      if (providerRef.current?.baseUrl && providerRef.current.baseUrl !== "auto") {
+        try {
+          const providerModels = await listModels(providerRef.current.baseUrl, providerRef.current.apiKey || "");
+          if (providerModels.length > 0) {
+            setModelPicker(providerModels);
+            setModelPickerIndex(0);
+            return;
+          }
+        } catch (err) {
+          // Provider fetch failed
         }
-      } catch {}
-      addMsg("info", `Current model: ${modelName}\n  Usage: /model <model-name>`);
+      }
+      
+      // No models found anywhere
+      addMsg("error", "No models available. Download one with /ollama pull or configure a provider.");
       return;
     }
     if (trimmed.startsWith("/model ")) {