llm-advanced-tools 0.1.0 → 0.1.1

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 hetpatel-11
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -8,9 +8,10 @@ A TypeScript library that brings Anthropic's advanced tool use features to all m
 Dynamically discover and load tools on-demand instead of loading everything upfront.
 
 **Benefits:**
-- 85%+ reduction in token usage
+- Reduces token usage by deferring tool loading
 - Improved accuracy with large tool sets
 - Scale to hundreds or thousands of tools
+- *Anthropic reports 85%+ token reduction in their testing*
 
 ### 🚀 Programmatic Tool Calling
 Orchestrate tools through code execution rather than individual API calls.
@@ -19,26 +20,26 @@ Orchestrate tools through code execution rather than individual API calls.
 - Keep intermediate results out of LLM context
 - Parallel tool execution
 - Better control flow with loops, conditionals, data transformations
-- 37%+ reduction in token usage on complex tasks
+- *Anthropic reports 37%+ token reduction on complex tasks in their testing*
 
 ### 📝 Tool Use Examples
 Provide sample invocations to improve tool call accuracy.
 
 **Benefits:**
-- 18%+ accuracy improvement on complex parameters
 - Show proper usage patterns
 - Clarify format conventions and optional parameters
+- *Anthropic reports 18%+ accuracy improvement on complex parameters in their testing*
 
 ## Installation
 
 ```bash
-npm install advanced-tools
+npm install llm-advanced-tools
 ```
 
 ## Quick Start
 
 ```typescript
-import { Client, ToolRegistry, OpenAIAdapter, ToolDefinition } from 'advanced-tools';
+import { Client, ToolRegistry, OpenAIAdapter, ToolDefinition } from 'llm-advanced-tools';
 
 // 1. Create a tool registry
 const registry = new ToolRegistry({
@@ -92,7 +93,7 @@ console.log(response);
 Use the latest Vercel AI SDK with any provider:
 
 ```typescript
-import { Client, ToolRegistry, VercelAIAdapter } from 'advanced-tools';
+import { Client, ToolRegistry, VercelAIAdapter } from 'llm-advanced-tools';
 import { openai } from '@ai-sdk/openai';
 import { anthropic } from '@ai-sdk/anthropic';
 
@@ -133,7 +134,7 @@ const response = await openaiClient.ask("What's the weather?");
                           │
                           ▼
 ┌─────────────────────────────────────────────────────────┐
-│              Unified Tool Interface                      │
+│              Unified Tool Interface                     │
 │  • ToolRegistry (search, defer loading)                 │
 │  • CodeExecutor (programmatic calling)                  │
 │  • ToolDefinition (with examples)                       │
@@ -141,7 +142,7 @@ const response = await openaiClient.ask("What's the weather?");
                           │
                           ▼
 ┌─────────────────────────────────────────────────────────┐
-│              Provider Adapters                           │
+│              Provider Adapters                          │
 │  ┌──────────┬──────────┬──────────┬──────────┐          │
 │  │ OpenAI   │Anthropic │ Google   │ Ollama   │          │
 │  └──────────┴──────────┴──────────┴──────────┘          │
@@ -223,7 +224,6 @@ The LLM learns proper usage patterns from the examples.
 | ↳ Google | ❌ (emulated) | ❌ (emulated) | ❌ (emulated) | ✅ via AI SDK |
 | ↳ Others | ❌ (emulated) | ❌ (emulated) | ❌ (emulated) | ✅ via AI SDK |
 | OpenAI (direct) | ❌ (emulated) | ❌ (emulated) | ❌ (emulated) | ✅ Ready |
-| Anthropic (direct) | ✅ | ✅ | ✅ | 🚧 Coming |
 | Ollama | ❌ (emulated) | ❌ (emulated) | ❌ (emulated) | 🚧 Coming |
 
 ## Configuration
@@ -311,17 +311,6 @@ class Client {
 }
 ```
 
-## Performance
-
-Based on internal testing:
-
-| Metric | Traditional | With Advanced Tools | Improvement |
-|--------|------------|---------------------|-------------|
-| Token usage (large tool set) | ~77K tokens | ~8.7K tokens | **88% reduction** |
-| Token usage (complex workflows) | 43,588 tokens | 27,297 tokens | **37% reduction** |
-| Tool selection accuracy | 79.5% | 88.1% | **+8.6 points** |
-| Parameter accuracy | 72% | 90% | **+18 points** |
-
 ## When to Use Each Feature
 
 ### Tool Search Tool
@@ -376,7 +365,6 @@ The default VM executor is **NOT secure** for untrusted code. For production:
 ## Roadmap
 
 - [x] Core library with OpenAI adapter
-- [ ] Anthropic adapter (native features)
 - [ ] Google/Gemini adapter
 - [ ] Ollama adapter with model capability detection
 - [ ] Docker-based executor
@@ -397,6 +385,6 @@ MIT
 ## Credits
 
 This library implements features described in Anthropic's blog post:
-[Introducing advanced tool use on the Claude Developer Platform](https://www.anthropic.com/news/advanced-tool-use)
+[Introducing advanced tool use on the Claude Developer Platform](https://www.anthropic.com/engineering/advanced-tool-use)
 
 The implementation is provider-agnostic and works with any LLM that supports function calling.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llm-advanced-tools",
-  "version": "0.1.0",
+  "version": "0.1.1",
   "description": "Provider-agnostic advanced tool use library for LLMs",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
@@ -29,19 +29,35 @@
   "license": "MIT",
   "dependencies": {
     "openai": "^4.0.0",
-    "@anthropic-ai/sdk": "^0.27.0",
+    "natural": "^6.0.0",
+    "zod": "^3.22.0"
+  },
+  "peerDependencies": {
     "ai": "^4.0.0",
     "@ai-sdk/openai": "^1.0.0",
-    "@ai-sdk/anthropic": "^1.0.0",
-    "zod": "^3.22.0",
-    "natural": "^6.0.0"
+    "@ai-sdk/anthropic": "^1.0.0"
+  },
+  "peerDependenciesMeta": {
+    "ai": {
+      "optional": true
+    },
+    "@ai-sdk/openai": {
+      "optional": true
+    },
+    "@ai-sdk/anthropic": {
+      "optional": true
+    }
   },
   "devDependencies": {
-    "@types/node": "^20.0.0",
+    "@ai-sdk/anthropic": "^1.0.0",
+    "@ai-sdk/openai": "^1.0.0",
+    "@anthropic-ai/sdk": "^0.27.0",
     "@types/jest": "^29.0.0",
     "@types/natural": "^5.1.0",
+    "@types/node": "^20.0.0",
     "@typescript-eslint/eslint-plugin": "^6.0.0",
     "@typescript-eslint/parser": "^6.0.0",
+    "ai": "^4.0.0",
     "eslint": "^8.0.0",
     "jest": "^29.0.0",
     "prettier": "^3.0.0",