@proveanything/smartlinks 2.0.49 → 2.0.50

package/dist/docs/API_SUMMARY.md

@@ -1,6 +1,6 @@
 # Smartlinks API Summary
 
-Version: 2.0.49  |  Generated: 2026-10-09T10:14:16.325Z
+Version: 2.0.50  |  Generated: 2026-10-09T10:28:38.661Z
 
 This is a concise summary of all available API functions and types.
 

package/dist/docs/ai.md

@@ -408,15 +408,21 @@ Discover tools two ways:
 
 ### Recommended Models
 
-For agentic workflows on `v1/responses`, GPT-5.6 ships in three tiers. Pass either the full model
-id (`openai/gpt-5.6-sol`) or the shorthand tier alias (`'cheap'`, `'balanced'`, `'premium'`) as
-`model` — both resolve through the same server-side model registry.
-
-| Tier | Model | Alias | Use for |
-|------|-------|-------|---------|
-| Cheapest (default) | `openai/gpt-5.6-luna` | `'cheap'` / omit `model` | Everyday chat, high-volume/simple turns |
-| Balanced | `openai/gpt-5.6-terra` | `'balanced'` | Admin agent / tool-heavy setup workflows (route default) |
-| Flagship | `openai/gpt-5.6-sol` | `'premium'` | Most demanding reasoning, coding, and multi-step tool use |
+Pass either a full model id (`openai/gpt-6.1-sol`) or a tier alias (`'cheap'`, `'balanced'`,
+`'premium'`) as `model` — both resolve through the same server-side model registry. The tiers are the
+GPT-6 family (prices per 1M tokens, input / output, OpenAI list price):
+
+| Tier | Model | Alias | Price | Use for |
+|------|-------|-------|-------|---------|
+| Cheapest (default) | `openai/gpt-6-luna` | `'cheap'` / omit `model` | $0.10 / $0.50 | Everyday chat, high-volume/simple turns |
+| Balanced | `openai/gpt-6.1-sol` | `'balanced'` | $2 / $10 | Admin agent / tool-heavy workflows (the `v1/responses` default) |
+| Most capable | `openai/gpt-6-astra` | `'premium'` | $10 / $50 | The hardest reasoning, coding and long multi-step tool use |
+
+The previous generation stays available by id — `openai/gpt-5.6-luna` ($0.20 / $1.20),
+`openai/gpt-5.6-terra` ($2 / $12) and `openai/gpt-5.6-sol` ($4 / $20) — and `ai.models.list(collectionId)` returns
+the full current set. GPT-6.1 Sol is the same input price as GPT-5.6 Terra with cheaper output, so prefer
+the tiers. If OpenAI refuses a GPT-6 request for a model/option reason, SmartLinks retries it once on the
+GPT-5.6 equivalent, so a call never fails just because a GPT-6 model doesn't accept something yet.
 
 Every tier supports function/tool calling, Programmatic Tool Calling, and Multi-agent (see below), and all
 three accept `service_tier: 'flex'` (≈50% cheaper, Batch-API rates, slower) or `'priority'` (premium, guaranteed

package/dist/types/ai.d.ts

@@ -679,7 +679,7 @@ export interface AgentRunRequest {
     prompt?: string;
     /** System instructions. */
     instructions?: string;
-    /** Model id, e.g. 'openai/gpt-5.6-terra'. */
+    /** Model id, e.g. 'openai/gpt-6.1-sol'. */
     model?: string;
     /** Safety cap on model round-trips. */
     maxSteps?: number;

package/docs/API_SUMMARY.md

@@ -1,6 +1,6 @@
 # Smartlinks API Summary
 
-Version: 2.0.49  |  Generated: 2026-10-09T10:14:16.325Z
+Version: 2.0.50  |  Generated: 2026-10-09T10:28:38.661Z
 
 This is a concise summary of all available API functions and types.
 

package/docs/ai.md

@@ -408,15 +408,21 @@ Discover tools two ways:
 
 ### Recommended Models
 
-For agentic workflows on `v1/responses`, GPT-5.6 ships in three tiers. Pass either the full model
-id (`openai/gpt-5.6-sol`) or the shorthand tier alias (`'cheap'`, `'balanced'`, `'premium'`) as
-`model` — both resolve through the same server-side model registry.
-
-| Tier | Model | Alias | Use for |
-|------|-------|-------|---------|
-| Cheapest (default) | `openai/gpt-5.6-luna` | `'cheap'` / omit `model` | Everyday chat, high-volume/simple turns |
-| Balanced | `openai/gpt-5.6-terra` | `'balanced'` | Admin agent / tool-heavy setup workflows (route default) |
-| Flagship | `openai/gpt-5.6-sol` | `'premium'` | Most demanding reasoning, coding, and multi-step tool use |
+Pass either a full model id (`openai/gpt-6.1-sol`) or a tier alias (`'cheap'`, `'balanced'`,
+`'premium'`) as `model` — both resolve through the same server-side model registry. The tiers are the
+GPT-6 family (prices per 1M tokens, input / output, OpenAI list price):
+
+| Tier | Model | Alias | Price | Use for |
+|------|-------|-------|-------|---------|
+| Cheapest (default) | `openai/gpt-6-luna` | `'cheap'` / omit `model` | $0.10 / $0.50 | Everyday chat, high-volume/simple turns |
+| Balanced | `openai/gpt-6.1-sol` | `'balanced'` | $2 / $10 | Admin agent / tool-heavy workflows (the `v1/responses` default) |
+| Most capable | `openai/gpt-6-astra` | `'premium'` | $10 / $50 | The hardest reasoning, coding and long multi-step tool use |
+
+The previous generation stays available by id — `openai/gpt-5.6-luna` ($0.20 / $1.20),
+`openai/gpt-5.6-terra` ($2 / $12) and `openai/gpt-5.6-sol` ($4 / $20) — and `ai.models.list(collectionId)` returns
+the full current set. GPT-6.1 Sol is the same input price as GPT-5.6 Terra with cheaper output, so prefer
+the tiers. If OpenAI refuses a GPT-6 request for a model/option reason, SmartLinks retries it once on the
+GPT-5.6 equivalent, so a call never fails just because a GPT-6 model doesn't accept something yet.
 
 Every tier supports function/tool calling, Programmatic Tool Calling, and Multi-agent (see below), and all
 three accept `service_tier: 'flex'` (≈50% cheaper, Batch-API rates, slower) or `'priority'` (premium, guaranteed

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@proveanything/smartlinks",
-  "version": "2.0.49",
+  "version": "2.0.50",
   "description": "Official JavaScript/TypeScript SDK for the Smartlinks API",
   "type": "module",
   "main": "dist/index.js",

package/scripts/doctor.mjs

@@ -163,6 +163,9 @@ if (meta.cssBaseline) {
       warnings.push(`meta.cssBaseline is "${meta.cssBaseline}" but this SDK ships baseline "${baseline.version}" — the host may serve a different set.`);
     }
     const baseSet = new Set(baseline.classes || []);
+    // SDK-defined names that aren't baseline helpers but are legitimately in bundles: the Sidekick
+    // screen's scope class (mobile-admin-container.md → Styling).
+    for (const c of ['sl-mobile-admin']) baseSet.add(c);
     // Class usage lives in compiled markup strings — present in every bundle regardless
     // of module format, so scan both umd + esm across all surfaces.
     const used = new Set();