Add the harness, envs, per-bot models and chat attachments

- Harness: providers, model catalog synced hourly, automatic free picks,
  cooldowns, tool loop with context trimming, MCP computer tools
- Bots queue runs replies and tasks through the harness; jobs wait
  (attempt refunded) while every model is resting
- Envs: Settings > Environment, scoped general/workspace/computer,
  written to computers and redacted from tool output
- Models: Settings > Models, per-harness provider select defaulting to
  OpenRouter, and a models modal on each bot's chat (the coordinator's
  edits the defaults)
- Chat: thinking row while a reply runs, file attachments (upload,
  paste, drop) that replies read as images, audio and document text
- Fix media uploads failing after a user's first file, and private
  files being served from the public media path

Co-Authored-By: Claude Opus 5.5 <[email protected]>

src/bots-queue/run-bot-job.ts

@@ -1,14 +1,42 @@
-import type { BotJobRunner } from "./types";
+import grabBotById from "@/src/functions/backend/db/bots/grab-bot-by-id";
+import HarnessError from "@/src/harness/harness-error";
 import BotJobFatalError from "./bot-job-fatal-error";
+import runBotReply from "./run-bot-reply";
+import runBotTask from "./run-bot-task";
+import type { BotJobRunner } from "./types";
 
 /**
- * Does a job's actual work: the bot follows `job.instructions` with its
- * full prompt (`grabBotPrompt`), on `job.computer_id` if set, and returns
- * the message for the user. This is where the model layer plugs in; until
- * it exists every job fails at once and the user is told.
+ * Does a job's actual work on the harness: a `reply` answers the user in
+ * chat (`runBotReply`), a `task` works through its brief (`runBotTask`).
+ *
+ * Harness failures that waiting can fix (rate limits, outages) are thrown
+ * as they are, so the queue retries with backoff; the rest are fatal.
  */
-const runBotJob: BotJobRunner = async () => {
-    throw new BotJobFatalError(`Bots can't run queued work yet.`);
+const runBotJob: BotJobRunner = async (context) => {
+    const { job } = context;
+    const bot = job.bot_id ? await grabBotById({ bot_id: job.bot_id }) : null;
+
+    if (!bot?.id || !bot.user_id) {
+        throw new BotJobFatalError(`The bot for this job no longer exists.`);
+    }
+
+    const owned = { ...bot, id: bot.id, user_id: bot.user_id };
+
+    try {
+        return job.kind == "reply"
+            ? await runBotReply({ ...context, bot: owned })
+            : await runBotTask({ ...context, bot: owned });
+    } catch (error: unknown) {
+        if (
+            error instanceof HarnessError &&
+            error.code != "aborted" &&
+            !error.retryable
+        ) {
+            throw new BotJobFatalError(error.message);
+        }
+
+        throw error;
+    }
 };
 
 export default runBotJob;