documentation updates; AI classes renamed

We now have AiApi, OllamaAiApi, and OpenAiApi. Documentation updates to
provide a bit more high-level clarity that was originally generated by
the agent.

AGENTS.md

@@ -43,7 +43,7 @@ pnpm --filter gadget-drone dev
 ## TypeScript Strictness
 
 | Package      | Strictness                                                                         |
-|---|---|
+| ------------ | ---------------------------------------------------------------------------------- |
 | `@gadget/ai` | `strict: true`                                                                     |
 | gadget-drone | `strict: true`                                                                     |
 | gadget-code  | `strict: true`, `noUnusedLocals`, `noUnusedParameters`, `noUncheckedIndexedAccess` |
@@ -52,7 +52,7 @@ pnpm --filter gadget-drone dev
 
 When adding a new feature or service, determine its scope:
 
-- **Shared concern** (AI, logging, config schema) → goes in `@gadget/ai`
+- **Shared AI concern** (AI logging, config schema) → goes in `@gadget/ai`
 - **Drone-only** (Bull queue, workspace file operations) → goes in gadget-drone
 - **Web-only** (Express routes, Mongoose models, session management) → goes in gadget-code
 

README.md

@@ -5,7 +5,7 @@ A self-hosted **Agentic Engineering Platform (AEP)** — an IDE that drives auto
 ## Projects
 
 | Package        | Role                                                                     |
-|---|---|
+| -------------- | ------------------------------------------------------------------------ |
 | `gadget-code`  | Web service — agentic IDE, browser UI, API server                        |
 | `gadget-drone` | Worker process — runs the agentic workflow loop in workspace directories |
 | `@gadget/ai`   | Shared AI API abstraction — Ollama and OpenAI, called by both            |
@@ -43,6 +43,26 @@ pnpm --filter gadget-drone dev
 
 ## Architecture
 
-gadget-code runs on server infrastructure (MongoDB, Redis, etc.) and serves the browser-based IDE. gadget-drone runs on end-user machines, connecting via WebSocket to gadget-code, and executes the agentic workflow loop against local project directories via remote control. gadget-drone never connects directly to MongoDB or Redis — it communicates entirely through the Gadget Code API.
+### @gadget/ai
 
-AI calls are handled by `@gadget/ai`, which both projects depend on. This keeps all AI SDK knowledge in one place.
+AI API calls are handled by `@gadget/ai`, which both projects depend on. This keeps all AI SDK knowledge in one place, and currently implements:
+
+- [AiApi](./packages/ai/src/api.ts) - abstract base class for all AI APIs/SDKs
+- [OllamaAiApi](./packages/ai/src/ollama.ts) - Ollama API implementation
+- [OpenAiApi](./packages/ai/src/openai.ts) - OpenAI API implementation
+
+### gadget-drone
+
+gadget-drone is a headless process runs on end-user machines, connecting via Socket.IO to gadget-code, to receive and execute work orders for the agentic workflow loop.
+
+At startup, gadget-drone examines `process.cwd()` to determine if it's a workspace directory, and if so, starts a worker process that connects to gadget-code and waits for work orders. It processes work orders in project directories in the gadget-drone workspace directory, and communicates events, status, and results to the IDE via gadget-code's web services and Socket.IO.
+
+gadget-drone never connects directly to MongoDB or Redis — it communicates entirely through the Gadget Code API.
+
+### gadget-code
+
+gadget-code runs on server infrastructure (MongoDB, Redis, etc.) and serves the browser-based IDE. The IDE connects to gadget-code via Socket.IO to send and receive commands in chat sessions to gadget-drone.
+
+gadget-code can be stacked on a single host for local development, and can achieve significant scale on a single host. It can also be deployed in tiers, potentially made of clusters, and the web tier can be horizontally scaled for production use with high availability.
+
+Libraries such as [redis-adapter](https://github.com/socketio/socket.io-redis-adapter) and [redis-emitter](https://github.com/socketio/socket.io-redis-emitter) are used for message routing and distribution in the Socket.IO library, handling the real-time message routing among gadget-code, gadget-drone, and the IDE running in the browser.

packages/ai/README.md

@@ -53,8 +53,8 @@ console.log(result.stats.duration.text); // formatted, e.g. "00:00:02"
 
 Abstract base class. Currently implemented:
 
-- **`AiOllamaApi`** — Ollama provider
+- **`OllamaAiApi`** — Ollama provider
-- **`AiOpenAiApi`** — OpenAI provider (stubbed)
+- **`OpenAiApi`** — OpenAI provider (stubbed)
 
 #### `ai.generate(model, options, streamCallback?)`
 
@@ -83,7 +83,7 @@ Configured via `IAiProvider` with `sdk: "ollama"`. Uses the `ollama` npm package
 
 ### OpenAI
 
-Configured via `IAiProvider` with `sdk: "openai"`. Stubbed — `chat()` and `generate()` throw `"Not yet implemented"`. Implement by wiring the `openai` npm package following the same pattern as `AiOllamaApi`.
+Configured via `IAiProvider` with `sdk: "openai"`. Stubbed — `chat()` and `generate()` throw `"Not yet implemented"`. Implement by wiring the `openai` npm package following the same pattern as `OllamaAiApi`.
 
 ## Duration Formatting
 

packages/ai/src/index.ts

@@ -19,11 +19,11 @@ export {
   AiApi,
 } from "./api.js";
 
-export { AiOllamaApi } from "./ollama.js";
+export { OllamaAiApi } from "./ollama.js";
-export { AiOpenAiApi } from "./openai.js";
+export { OpenAiApi } from "./openai.js";
 
-import { AiOllamaApi } from "./ollama.js";
+import { OllamaAiApi } from "./ollama.js";
-import { AiOpenAiApi } from "./openai.js";
+import { OpenAiApi } from "./openai.js";
 import type { IAiProvider } from "./api.js";
 import type { IAiLogger } from "./api.js";
 import type { AiApi } from "./api.js";
@@ -31,9 +31,9 @@ import type { AiApi } from "./api.js";
 export function createAiApi(provider: IAiProvider, logger?: IAiLogger): AiApi {
   switch (provider.sdk) {
     case "ollama":
-      return new AiOllamaApi(provider, logger);
+      return new OllamaAiApi(provider, logger);
     case "openai":
-      return new AiOpenAiApi(provider, logger);
+      return new OpenAiApi(provider, logger);
     default:
       throw new Error(`Unknown AI SDK: ${(provider as IAiProvider).sdk}`);
   }

packages/ai/src/ollama.ts

@@ -18,7 +18,7 @@ import {
   IAiResponseStreamFn,
 } from "./api.js";
 
-export class AiOllamaApi extends AiApi {
+export class OllamaAiApi extends AiApi {
   protected client: Ollama;
 
   constructor(provider: IAiProvider, logger?: IAiLogger) {
@@ -30,11 +30,11 @@ export class AiOllamaApi extends AiApi {
   }
 
   async listModels(): Promise<void> {
-    await this.log.debug("AiOllamaApi.listModels called");
+    await this.log.debug("OllamaAiApi.listModels called");
   }
 
   async probeModel(modelId: string): Promise<void> {
-    await this.log.debug("AiOllamaApi.probeModel called", { modelId });
+    await this.log.debug("OllamaAiApi.probeModel called", { modelId });
   }
 
   async generate(
@@ -42,7 +42,7 @@ export class AiOllamaApi extends AiApi {
     options: IAiGenerateOptions,
     streamCallback?: IAiResponseStreamFn,
   ): Promise<IAiGenerateResponse> {
-    await this.log.debug("AiOllamaApi.generate called", {
+    await this.log.debug("OllamaAiApi.generate called", {
       provider: model.provider.name,
       modelId: model.modelId,
     });
@@ -85,7 +85,7 @@ export class AiOllamaApi extends AiApi {
     options: IAiChatOptions,
     streamCallback?: IAiResponseStreamFn,
   ): Promise<IAiChatResponse> {
-    await this.log.debug("AiOllamaApi.chat called", {
+    await this.log.debug("OllamaAiApi.chat called", {
       provider: model.provider.name,
       modelId: model.modelId,
     });

packages/ai/src/openai.ts

@@ -14,17 +14,17 @@ import {
   IAiResponseStreamFn,
 } from "./api.js";
 
-export class AiOpenAiApi extends AiApi {
+export class OpenAiApi extends AiApi {
   constructor(provider: IAiProvider, logger?: IAiLogger) {
     super(provider, logger);
   }
 
   async listModels(): Promise<void> {
-    await this.log.debug("AiOpenAiApi.listModels called");
+    await this.log.debug("OpenAiApi.listModels called");
   }
 
   async probeModel(modelId: string): Promise<void> {
-    await this.log.debug("AiOpenAiApi.probeModel called", { modelId });
+    await this.log.debug("OpenAiApi.probeModel called", { modelId });
   }
 
   async generate(