fix: require user to specify the provider

README.md

@@ -18,13 +18,17 @@ bun install # or use `mise install`
 
 ## Quick start
 
-**1. Create a config file**
+**1. Initialize workspace**
 
 ```bash
-mkdir -p ~/.config/nanobot
+bun run nanobot onboard
 ```
 
-`~/.config/nanobot/config.json`:
+This creates `~/.config/nanobot/` with a config file and templates.
+
+**2. Edit config**
+
+Add your API key and set provider/model:
 
 ```json
 {
@@ -34,12 +38,13 @@ mkdir -p ~/.config/nanobot
     }
   },
   "agent": {
-    "model": "openrouter/anthropic/claude-sonnet-4-5"
+    "provider": "openrouter",
+    "model": "anthropic/claude-sonnet-4-5"
   }
 }
 ```
 
-**2. Chat**
+**3. Chat**
 
 ```bash
 bun run nanobot agent
@@ -115,7 +120,8 @@ Environment variable overrides:
 ```json
 {
   "agent": {
-    "model": "openrouter/anthropic/claude-sonnet-4-5",
+    "provider": "openrouter",
+    "model": "anthropic/claude-sonnet-4-5",
     "workspacePath": "~/.config/nanobot",
     "maxTokens": 4096,
     "contextWindowTokens": 65536,
@@ -164,17 +170,27 @@ Environment variable overrides:
 }
 ```
 
-### Providers
+### Provider
 
-Model names use a `provider/model` prefix scheme:
+The `agent.provider` field is **required** and must be one of:
 
-| Prefix | Provider | Example |
-|--------|----------|---------|
-| `anthropic/` | Anthropic direct | `anthropic/claude-opus-4-5` |
-| `openai/` | OpenAI direct | `openai/gpt-4o` |
-| `google/` | Google direct | `google/gemini-2.5-pro` |
-| `openrouter/` | OpenRouter (any model) | `openrouter/anthropic/claude-sonnet-4-5` |
-| `ollama/` | Local Ollama | `ollama/llama3.2` |
+| Provider | Description |
+|----------|-------------|
+| `anthropic` | Anthropic direct (Claude models) |
+| `openai` | OpenAI direct (GPT models) |
+| `google` | Google direct (Gemini models) |
+| `openrouter` | OpenRouter (access to many models) |
+| `ollama` | Local Ollama instance |
+
+The `agent.model` field is also **required** and should be the model ID without any provider prefix:
+
+| Provider | Example Model |
+|----------|---------------|
+| `anthropic` | `claude-sonnet-4-5`, `claude-opus-4-5` |
+| `openai` | `gpt-4o`, `gpt-4o-mini` |
+| `google` | `gemini-2.5-pro`, `gemini-2.0-flash` |
+| `openrouter` | `anthropic/claude-sonnet-4-5` (OpenRouter uses its own model IDs) |
+| `ollama` | `llama3.2`, `qwen2.5` |
 
 For Ollama, set `providers.ollama.apiBase` (default: `http://localhost:11434/api`).
 

memory-bank/activeContext.md

@@ -19,13 +19,14 @@ Docs directory created with 4 files (PRD.md, Architecture.md, API.md, Discoverie
 - **lint**: all `${err}` in template literals → `${String(err)}`; `String(args['key'] ?? '')` → `strArg(args, 'key')` helper; unused `onProgress` param → `_onProgress`; WebSocket `onerror` `err` type is `Event` → use `err.type`
 
 ## Work Queue (next steps)
-1. [x] Create workspace helper module (src/workspace.ts) with ensureWorkspace() and syncTemplates()
+1. [x] Create workspace helper module (src/cli/utils.ts) with ensureWorkspace() and syncTemplates()
 2. [x] Create onboard command (src/cli/onboard.ts) with path argument and directory-not-empty guard
-3. [x] Update src/cli/commands.ts to use ensureWorkspace() instead of inline mkdirSync
-4. [x] Typecheck and lint pass (0 errors)
-5. [x] Runtime smoke test: `bun run nanobot --help`
-6. [x] Test onboard command: `bun run nanobot onboard [path]`
-7. [ ] Test with a real Mattermost config (optional — user can do this)
+3. [x] Agent/gateway commands check workspace exists (throw if not found)
+4. [x] Added required `provider` field to agent config (values: anthropic, openai, google, openrouter, ollama)
+5. [x] Provider resolution uses explicit provider from config (no model prefix parsing)
+6. [x] Typecheck and lint pass (0 errors)
+7. [x] Test onboard and agent commands work correctly
+8. [ ] Test with a real Mattermost config (optional — user can do this)
 
 ## Key Decisions Made
 - Mattermost channel uses raw WebSocket + fetch (no mattermostdriver, no SSL hack)
@@ -34,3 +35,5 @@ Docs directory created with 4 files (PRD.md, Architecture.md, API.md, Discoverie
 - Config is fresh Zod schema (no migration from Python config needed)
 - `ollama-ai-provider` package (not `@ai-sdk/ollama` which 404s on npm)
 - `strArg(args, key, fallback?)` helper exported from `agent/tools/base.ts` for safe unknown→string extraction
+- Agent config requires explicit `provider` field (no more model prefix like "anthropic/claude-...")
+- Model names are now just the raw model ID (e.g., "claude-sonnet-4-5" not "anthropic/claude-sonnet-4-5")

memory-bank/progress.md

@@ -8,49 +8,20 @@
 - All dependencies installed
 - `src/` directory structure scaffolded
 - Memory bank initialized
-- All source files written (first pass):
-  - `src/config/types.ts` + `src/config/loader.ts`
-  - `src/bus/types.ts` + `src/bus/queue.ts`
-  - `src/provider/types.ts` + `src/provider/index.ts`
-  - `src/session/types.ts` + `src/session/manager.ts`
-  - `src/agent/tools/base.ts` (+ `strArg` helper)
-  - `src/agent/tools/filesystem.ts`
-  - `src/agent/tools/shell.ts`
-  - `src/agent/tools/web.ts`
-  - `src/agent/tools/message.ts`
-  - `src/agent/tools/spawn.ts` + `src/agent/subagent.ts`
-  - `src/agent/tools/cron.ts`
-  - `src/cron/types.ts` + `src/cron/service.ts`
-  - `src/heartbeat/service.ts`
-  - `src/agent/memory.ts`
-  - `src/agent/skills.ts`
-  - `src/agent/context.ts`
-  - `src/agent/loop.ts`
-  - `src/channels/base.ts` + `src/channels/mattermost.ts`
-  - `src/channels/manager.ts`
-  - `src/cli/commands.ts`
-  - `index.ts`
+- All source files written (first pass)
 - Templates and skills copied from Python repo
 - **Full typecheck pass**: `tsc --noEmit` → 0 errors
 - **Full lint pass**: `oxlint` → 0 errors, 0 warnings
 - `package.json` scripts added: `start`, `dev`, `typecheck`
 - **Docs created**: `/docs/PRD.md`, `Architecture.md`, `API.md`, `Discoveries.md`
+- **Onboard command**: Created `src/cli/onboard.ts` with workspace initialization
+- **Provider config**: Added required `provider` field to agent config
+- **Workspace validation**: Agent/gateway commands throw if workspace doesn't exist
 
 ### 🔄 In Progress
 - Nothing
 
-### ✅ Done
-- Created src/workspace.ts with ensureWorkspace(), syncTemplates(), checkWorkspaceEmpty()
-- Created src/cli/onboard.ts command with path argument
-- Updated src/cli/commands.ts to use ensureWorkspace() helper
-- Typecheck: 0 errors
-- Lint: 0 warnings
-
-### 🔄 In Progress
-- Testing onboard command
-
 ### ⏳ Pending
-- Runtime smoke test: `bun run nanobot --help`
 - Integration test with a real Mattermost server
 
 ## Known Issues / Risks

src/cli/agent.ts

@@ -24,6 +24,7 @@ export function agentCommand(program: Command): void {
       const model = opts.model ?? config.agent.model;
       const provider = makeProvider(
         config.providers,
+        config.agent.provider,
         model,
         config.agent.maxTokens,
         config.agent.temperature,

src/cli/gateway.ts

@@ -24,6 +24,7 @@ export function gatewayCommand(program: Command): void {
 
       const provider = makeProvider(
         config.providers,
+        config.agent.provider,
         config.agent.model,
         config.agent.maxTokens,
         config.agent.temperature,

src/cli/onboard.ts

@@ -2,7 +2,6 @@ import { writeFileSync } from 'node:fs';
 import { join } from 'node:path';
 import { Command } from 'commander';
 import pc from 'picocolors';
-import { ConfigSchema, type Config } from '../config/types.ts';
 import { WORKSPACE_PATH } from '../config/constants.ts'
 import { ensureWorkspace, resolvePath, checkWorkspaceEmpty, syncTemplates } from './utils.ts';
 
@@ -16,12 +15,14 @@ export function onboardCommand(program: Command): void {
     .description('Initialize a new nanobot workspace with config and templates')
     .action(async (rawPath?: string) => {
       try {
-        const defaultConfig: Config = ConfigSchema.parse({
+        // Create a minimal config template - users must fill in provider and model
+        const defaultConfig = {
+          providers: {},
           agent: {
             provider: '',
             model: '',
-          }
-        });
+          },
+        };
 
         const targetPath = resolvePath(rawPath ?? WORKSPACE_PATH);
         const configPath = join(targetPath, 'config.json');

src/config/loader.ts

@@ -3,7 +3,6 @@ import { homedir } from 'node:os';
 import { dirname, resolve } from 'node:path';
 import pc from 'picocolors'
 import { type Config, ConfigSchema } from './types.ts';
-import { WORKSPACE_PATH } from './constants.ts'
 
 const DEFAULT_CONFIG_PATH = resolve(homedir(), '.config', 'nanobot', 'config.json');
 

src/config/types.ts

@@ -40,8 +40,18 @@ export type ChannelsConfig = z.infer<typeof ChannelsConfigSchema>;
 // Agent
 // ---------------------------------------------------------------------------
 
+export const AgentProviderSchema = z.enum([
+  'anthropic',
+  'openai',
+  'google',
+  'openrouter',
+  'ollama',
+]);
+export type AgentProvider = z.infer<typeof AgentProviderSchema>;
+
 export const AgentConfigSchema = z.object({
-  model: z.string().default('anthropic/claude-sonnet-4-5'),
+  provider: AgentProviderSchema,
+  model: z.string(),
   workspacePath: z.string().default(WORKSPACE_PATH),
   maxTokens: z.number().int().default(4096),
   contextWindowTokens: z.number().int().default(65536),

src/provider/index.ts

@@ -5,7 +5,7 @@ import { createOpenRouter } from '@openrouter/ai-sdk-provider';
 import { type ModelMessage, generateText, stepCountIs } from 'ai';
 import { jsonrepair } from 'jsonrepair';
 import { createOllama } from 'ollama-ai-provider';
-import type { ProvidersConfig } from '../config/types.ts';
+import type { AgentProvider, ProvidersConfig } from '../config/types.ts';
 import type { ChatOptions, LLMResponse, ToolDefinition } from './types.ts';
 
 export type { ToolDefinition };
@@ -66,17 +66,20 @@ import type { LanguageModel } from 'ai';
 
 export class LLMProvider {
   private _providers: ProvidersConfig;
+  private _provider: AgentProvider;
   private _defaultModel: string;
   private _maxTokens: number;
   private _temperature: number;
 
   constructor(
     providers: ProvidersConfig,
+    provider: AgentProvider,
     defaultModel: string,
     maxTokens = 4096,
     temperature = 0.7,
   ) {
     this._providers = providers;
+    this._provider = provider;
     this._defaultModel = defaultModel;
     this._maxTokens = maxTokens;
     this._temperature = temperature;
@@ -87,39 +90,30 @@ export class LLMProvider {
   }
 
   private _resolveModel(model: string): LanguageModel {
-    const slashIdx = model.indexOf('/');
-    const prefix = slashIdx >= 0 ? model.slice(0, slashIdx) : model;
-    const remainder = slashIdx >= 0 ? model.slice(slashIdx + 1) : model;
-
-    switch (prefix) {
+    switch (this._provider) {
       case 'anthropic': {
         const cfg = this._providers.anthropic;
-        return createAnthropic({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(remainder);
+        return createAnthropic({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(model);
       }
       case 'openai': {
         const cfg = this._providers.openai;
-        return createOpenAI({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(remainder);
+        return createOpenAI({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(model);
       }
       case 'google': {
         const cfg = this._providers.google;
-        return createGoogleGenerativeAI({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(remainder);
+        return createGoogleGenerativeAI({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(model);
       }
       case 'openrouter': {
         const cfg = this._providers.openrouter;
-        return createOpenRouter({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(remainder);
+        return createOpenRouter({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(model);
       }
       case 'ollama': {
         const cfg = this._providers.ollama;
         // ollama-ai-provider returns LanguageModelV1; cast to LanguageModel (compatible at runtime)
         return createOllama({ baseURL: cfg?.apiBase ?? 'http://localhost:11434/api' })(
-          remainder,
+          model,
         ) as unknown as LanguageModel;
       }
-      default: {
-        // No recognized prefix — fall through to openai-compatible
-        const cfg = this._providers.openai;
-        return createOpenAI({ apiKey: cfg?.apiKey, baseURL: cfg?.apiBase })(model);
-      }
     }
   }
 
@@ -212,11 +206,12 @@ export class LLMProvider {
 
 export function makeProvider(
   providers: ProvidersConfig,
+  provider: AgentProvider,
   model: string,
   maxTokens: number,
   temperature: number,
 ): LLMProvider {
-  return new LLMProvider(providers, model, maxTokens, temperature);
+  return new LLMProvider(providers, provider, model, maxTokens, temperature);
 }
 
 /** Build a tool-result message to append after executing a tool call. */