@continum/sdk 0.0.4 → 0.0.6

package/README.md

@@ -2,6 +2,13 @@
 
 > Governed Execution Framework for LLM Applications
 
+**Current Version**: 0.0.5
+
+**Latest Changes**:
+- Fixed `max_tokens` validation for Anthropic thinking models
+- Default `max_tokens` increased to 16000 for thinking models (Opus, Sonnet 4)
+- Better error messages when `max_tokens` is too low for thinking budget
+
 ## Quick Start
 
 ### Installation
@@ -24,20 +31,77 @@ import { Continum } from '@continum/sdk';
 const continum = new Continum({
   continumKey: process.env.CONTINUM_KEY!,
   apiKeys: {
-    openai: process.env.OPENAI_API_KEY
+    openai: process.env.OPENAI_API_KEY,
+    anthropic: process.env.ANTHROPIC_API_KEY,
+    gemini: process.env.GEMINI_API_KEY
   },
   defaultSandbox: 'pii_protection'
 });
 
-// Make LLM calls with automatic compliance
+// OpenAI - use snake_case model names
 const response = await continum.llm.openai.gpt_4o.chat({
   messages: [{ role: 'user', content: 'Hello world' }]
 });
+
+// Anthropic - use model family names (opus, sonnet, haiku)
+const response2 = await continum.llm.anthropic.opus_4_6.chat({
+  messages: [{ role: 'user', content: 'Review this code' }]
+});
+// Also supports: sonnet_4_6, sonnet_4, haiku_4_5, haiku_3_5, sonnet_3_7
+// Legacy format also works: claude_3_5_sonnet
+
+// Gemini - use snake_case with underscores
+const response3 = await continum.llm.gemini.gemini_2_5_pro.chat({
+  messages: [{ role: 'user', content: 'Summarize this' }]
+});
+
 // ✅ Guardian checks for PII (pre-execution)
 // ✅ User gets response instantly
 // ✅ Shadow Audit runs in background (post-execution)
 ```
 
+## Model Name Format
+
+The SDK uses snake_case model names that get automatically transformed to the correct API format:
+
+### Anthropic Models
+
+```typescript
+// Recommended format - use model family names
+continum.llm.anthropic.opus_4_6.chat()      // → claude-opus-4-6
+continum.llm.anthropic.sonnet_4_6.chat()    // → claude-sonnet-4-6
+continum.llm.anthropic.sonnet_4.chat()      // → claude-sonnet-4-5
+continum.llm.anthropic.haiku_4_5.chat()     // → claude-haiku-4-5-20251001
+continum.llm.anthropic.haiku_3_5.chat()     // → claude-haiku-3-5-20241022
+continum.llm.anthropic.sonnet_3_7.chat()    // → claude-sonnet-3-7-20250219
+
+// Legacy format also supported (v0.0.4+)
+continum.llm.anthropic.claude_3_5_sonnet.chat()  // → claude-3-5-sonnet-20241022
+
+// Alias: claude and anthropic are interchangeable
+continum.llm.claude.opus_4_6.chat()  // Same as anthropic.opus_4_6
+```
+
+### OpenAI Models
+
+```typescript
+continum.llm.openai.gpt_5.chat()       // → gpt-5
+continum.llm.openai.gpt_4o.chat()      // → gpt-4o
+continum.llm.openai.gpt_4_turbo.chat() // → gpt-4-turbo
+continum.llm.openai.o3.chat()          // → o3
+continum.llm.openai.o3_mini.chat()     // → o3-mini
+continum.llm.openai.o1.chat()          // → o1
+```
+
+### Gemini Models
+
+```typescript
+continum.llm.gemini.gemini_2_5_pro.chat()   // → gemini-2.5-pro
+continum.llm.gemini.gemini_2_5_flash.chat() // → gemini-2.5-flash
+continum.llm.gemini.gemini_2_0_flash.chat() // → gemini-2.0-flash
+continum.llm.gemini.gemini_1_5_pro.chat()   // → gemini-1.5-pro
+```
+
 ## Architecture
 
 ### Two-Tier Protection System

package/dist/drivers/anthropic.driver.js

@@ -7,9 +7,17 @@ class AnthropicDriver {
     }
     async call(model, params) {
         const isThinkingModel = model.includes('opus') || model.includes('sonnet-4') || model.includes('claude-4');
+        const thinkingBudget = 8000;
+        const defaultMaxTokens = isThinkingModel ? 16000 : 4096;
+        const maxTokens = params.maxTokens ?? defaultMaxTokens;
+        // Ensure max_tokens is greater than thinking budget
+        if (isThinkingModel && maxTokens <= thinkingBudget) {
+            throw new Error(`max_tokens (${maxTokens}) must be greater than thinking.budget_tokens (${thinkingBudget}). ` +
+                `Please set maxTokens to at least ${thinkingBudget + 1} or use the default.`);
+        }
         const body = {
             model,
-            max_tokens: params.maxTokens ?? 4096,
+            max_tokens: maxTokens,
             messages: params.messages.map(m => ({ role: m.role, content: m.content })),
         };
         if (params.systemPrompt) {
@@ -17,7 +25,7 @@ class AnthropicDriver {
         }
         // Enable extended thinking for capable models
         if (isThinkingModel) {
-            body.thinking = { type: 'enabled', budget_tokens: 8000 };
+            body.thinking = { type: 'enabled', budget_tokens: thinkingBudget };
         }
         else {
             body.temperature = params.temperature ?? 0.7;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@continum/sdk",
-  "version": "0.0.4",
+  "version": "0.0.6",
   "description": "Zero-latency compliance auditing for every LLM call in your application",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",