@columenlabs/lumen-ai 0.1.144 → 0.1.146

package/README.md

@@ -102,6 +102,7 @@ app.all('/api/assistant/*', (c) => routes(c.req.raw));
 - **Web.** `fetch` is a plain GET from the Worker (10 s, 2 MB, never the app's own hosts or private addresses). ScrapeQL's own tools join over its MCP API once an admin pastes a Cloudflare Access service token (`PUT /web/scrapeql`). A Worker on ScrapeQL's own Cloudflare account passes its service binding instead, `scrapeql: { url: 'https://scrapeql/mcp', fetch: env.SCRAPEQL.fetch.bind(env.SCRAPEQL) }`: no token is stored, the tools are always on and Settings shows the link with a Test.
 - **Hand off and advice.** The `escalate` tool hands one task to a higher tier (the person sets next, any or only when asked); Anthropic's advisor tool lets a higher tier send a plan, at most two an answer.
 - **Chats** stream as server-sent events (`POST /chat`); closing the request stops the answer and keeps what was said. Past 150k tokens, or on `/compact`, the history sent to the model becomes a summary. Run `assistant.purge()` once a day to drop chats 30 days after their last message.
+- **Background work** an app runs for a person (a score, a summary) is `assistant.complete(caller, { task, system, prompt, maxTokens })`: one answer on the task's tier (`tasks`, so the person picks its model in Settings ▸ Models), no chat and no tools, billed to their payer and counted in their spend and allowance like a chat turn. It throws `AssistantError` `not_connected`, `app_cap` or `own_cap` when nobody can pay, so the caller waits and tries later.
 
 ### Commands, `@` and `#`
 

package/dist/assistant/engine.d.ts

@@ -241,6 +241,23 @@ export declare class Assistant {
     private loopClaude;
     /** Swaps the history for a short summary. False with nothing to compact. */
     private compact;
+    /**
+     * One prompt answered on a task's tier as the person: no chat, no tools. It is for background
+     * work an app runs for someone (a job's fit score, a summary), billed to their payer and capped
+     * like a chat turn, so it shows in their spend and allowance. Throws `AssistantError` with code
+     * `not_connected`, `app_cap` or `own_cap` when nobody can pay, or `upstream` when the model fails.
+     */
+    complete(caller: Caller, o: {
+        task: string;
+        system: string;
+        prompt: string;
+        maxTokens?: number;
+        signal?: AbortSignal;
+    }): Promise<{
+        text: string;
+        usd: number;
+        model: string;
+    }>;
     /** How the person's cards ended since the last answer, once each. */
     private cardNews;
     /** Starts a person's ChatGPT plan sign-in: the address to open in a new tab. */

package/dist/assistant/engine.js

@@ -1745,6 +1745,68 @@ export class Assistant {
         emit('compacted', {});
         return true;
     }
+    // ---------------------------------------------------------------- one-shot work
+    /**
+     * One prompt answered on a task's tier as the person: no chat, no tools. It is for background
+     * work an app runs for someone (a job's fit score, a summary), billed to their payer and capped
+     * like a chat turn, so it shows in their spend and allowance. Throws `AssistantError` with code
+     * `not_connected`, `app_cap` or `own_cap` when nobody can pay, or `upstream` when the model fails.
+     */
+    async complete(caller, o) {
+        const s = await this.session(caller, await this.mainTier(caller, o.task));
+        const { paying, wait } = await this.state.paying(caller.id, s.model.provider);
+        if (!paying) {
+            const code = wait ?? 'not_connected';
+            throw new AssistantError(waitText(code, s.model.provider, !!this.o.allowance), code, 402);
+        }
+        const clients = this.clients(paying);
+        const max = Math.min(o.maxTokens ?? 1000, 8000);
+        let text = '';
+        try {
+            if (s.model.provider === 'openai' && clients.openai) {
+                const res = await clients.openai.responses.create({
+                    model: s.model.id,
+                    instructions: o.system,
+                    input: o.prompt,
+                    store: false,
+                    max_output_tokens: max,
+                }, o.signal ? { signal: o.signal } : undefined);
+                const u = res.usage;
+                if (u)
+                    s.meter.add({
+                        input_tokens: u.input_tokens - (u.input_tokens_details?.cached_tokens ?? 0),
+                        output_tokens: u.output_tokens,
+                        cache_read_input_tokens: u.input_tokens_details?.cached_tokens ?? 0,
+                    }, !!clients.openaiPlan);
+                text = res.output_text ?? '';
+            }
+            else if (clients.anthropic) {
+                const msg = await clients.anthropic.messages.create({
+                    model: s.model.id,
+                    max_tokens: max,
+                    system: o.system,
+                    messages: [{ role: 'user', content: o.prompt }],
+                    ...effortParams(s.effort),
+                }, o.signal ? { signal: o.signal } : undefined);
+                s.meter.add(msg.usage);
+                text = msg.content
+                    .filter((b) => b.type === 'text')
+                    .map((b) => b.text)
+                    .join('');
+            }
+            else
+                throw new AssistantError(waitText('not_connected', s.model.provider), 'not_connected', 402);
+        }
+        catch (e) {
+            if (e instanceof AssistantError)
+                throw e;
+            throw new AssistantError(`${s.model.label} did not answer: ${String(e?.message ?? e).slice(0, 200)}`, 'upstream', 502);
+        }
+        finally {
+            await this.state.addSpend(caller.id, paying.payer, [s.meter.part()]);
+        }
+        return { text, usd: s.meter.usd, model: s.model.id };
+    }
     /** How the person's cards ended since the last answer, once each. */
     cardNews(doc) {
         const told = new Set(doc.told ?? []);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@columenlabs/lumen-ai",
-  "version": "0.1.144",
+  "version": "0.1.146",
   "description": "Lumen's assistant and agent parts: a chat assistant's server (Claude tiers, tools over MCP, the web, hand off, spend caps), serving an MCP server from a Worker, read-only tool sets and metered tool calls.",
   "license": "MIT",
   "type": "module",
@@ -41,8 +41,8 @@
   },
   "dependencies": {
     "@anthropic-ai/sdk": "0.133.0",
-    "@columenlabs/lumen-ops": "0.1.144",
-    "@columenlabs/lumen-ui-types": "0.1.144",
+    "@columenlabs/lumen-ops": "0.1.146",
+    "@columenlabs/lumen-ui-types": "0.1.146",
     "openai": "7.32.0",
     "zod": "^4.3.6"
   },