tooluniverse 1.0.0__py3-none-any.whl → 1.0.2__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.

This version of tooluniverse might be problematic. Click here for more details.

Files changed (20) hide show
  1. tooluniverse/__init__.py +71 -57
  2. tooluniverse/alphafold_tool.py +51 -20
  3. tooluniverse/compose_scripts/tool_graph_generation.py +249 -0
  4. tooluniverse/compose_scripts/tool_metadata_generator.py +369 -0
  5. tooluniverse/data/agentic_tools.json +143 -28
  6. tooluniverse/data/alphafold_tools.json +203 -61
  7. tooluniverse/data/compose_tools.json +63 -0
  8. tooluniverse/data/special_tools.json +2 -0
  9. tooluniverse/test/test_alphafold_tool.py +66 -29
  10. {tooluniverse-1.0.0.dist-info → tooluniverse-1.0.2.dist-info}/METADATA +115 -173
  11. {tooluniverse-1.0.0.dist-info → tooluniverse-1.0.2.dist-info}/RECORD +16 -18
  12. tooluniverse/data/packages/software_tools.json +0 -4954
  13. tooluniverse/remote/expert_feedback_mcp/human_expert_mcp_server.py +0 -1611
  14. tooluniverse/remote/expert_feedback_mcp/simple_test.py +0 -34
  15. tooluniverse/remote/expert_feedback_mcp/start_web_interface.py +0 -91
  16. /tooluniverse/data/{clait_tools.json → adverse_event_tools.json} +0 -0
  17. {tooluniverse-1.0.0.dist-info → tooluniverse-1.0.2.dist-info}/WHEEL +0 -0
  18. {tooluniverse-1.0.0.dist-info → tooluniverse-1.0.2.dist-info}/entry_points.txt +0 -0
  19. {tooluniverse-1.0.0.dist-info → tooluniverse-1.0.2.dist-info}/licenses/LICENSE +0 -0
  20. {tooluniverse-1.0.0.dist-info → tooluniverse-1.0.2.dist-info}/top_level.txt +0 -0
{tooluniverse-1.0.0.dist-info → tooluniverse-1.0.2.dist-info}/METADATA RENAMED
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: tooluniverse
3
- Version: 1.0.0
3
+ Version: 1.0.2
4
4
  Summary: A comprehensive collection of scientific tools for Agentic AI, offering integration with the ToolUniverse SDK and MCP Server to support advanced scientific workflows.
5
5
  Author-email: Shanghua Gao <shanghuagao@gmail.com>
6
6
  Project-URL: Homepage, https://github.com/mims-harvard/TxAgent
@@ -63,20 +63,49 @@ Provides-Extra: all
63
63
  Requires-Dist: tooluniverse[dev,docs,graph]; extra == "all"
64
64
  Dynamic: license-file
65
65
 
66
- # ToolUniverse
66
+ # ToolUniverse: Democratizing AI scientists
67
67
 
68
- [![ToolUniverse-PIP](https://img.shields.io/badge/Pip-ToolUniverse-blue)](https://pypi.org/project/tooluniverse/)
68
+ [![ToolUniverse-PIP](https://img.shields.io/badge/PyPI-ToolUniverse-blue)](https://pypi.org/project/tooluniverse/)
69
69
  [![ToolUniverse](https://img.shields.io/badge/Code-ToolUniverse-purple)](https://github.com/mims-harvard/ToolUniverse)
70
70
  [![Model context protocol (MCP)](https://img.shields.io/badge/Model_Context_Protocol_(MCP)_Supported-green)](README_USAGE.md#running-the-mcp-server)
71
71
  [![Documentation](https://img.shields.io/badge/Documentation-Available-green)](https://zitniklab.hms.harvard.edu/bioagent/)
72
- [![Web Service](https://img.shields.io/badge/Web_Service-aiscientist.tools-blue)](https://aiscientist.tools)
72
+ [![🌐Web Service](https://img.shields.io/badge/Web_Service-aiscientist.tools-blue)](https://aiscientist.tools)
73
+ [![Slack](https://img.shields.io/badge/Slack-Join_Community-orange)](https://join.slack.com/t/tooluniversehq/shared_invite/zt-3dic3eoio-5xxoJch7TLNibNQn5_AREQ)
73
74
 
74
- **Democratizing AI Scientists for Collaborative Discovery** 🌐 [Web Service](https://aiscientist.tools) | 📦 [PyPI](https://pypi.org/project/tooluniverse) | 📚 [Documentation](https://zitniklab.hms.harvard.edu/bioagent/)
75
75
 
76
76
 
77
+ **Democratizing AI scientists using ToolUniverse**
78
+
77
79
  ToolUniverse is an ecosystem for creating AI scientist systems from any open or closed large language model (LLM). It standardizes how LLMs identify and call tools, integrating more than **600 machine learning models, datasets, APIs, and scientific packages** for data analysis, knowledge retrieval, and experimental design.
78
80
 
79
- ## What is ToolUniverse?
81
+ <div align="center">
82
+
83
+ [![Visit Website](https://img.shields.io/badge/Visit_Website-AIScientist.Tools-4285F4?style=for-the-badge&logo=google-chrome&logoColor=white)](https://aiscientist.tools)
84
+
85
+ [![Join Slack Community](https://img.shields.io/badge/Join_Slack_Community-Get_Help_&_Connect-FF6B6B?style=for-the-badge&logo=slack&logoColor=white)](https://join.slack.com/t/tooluniversehq/shared_invite/zt-3dic3eoio-5xxoJch7TLNibNQn5_AREQ)
86
+ [![Join WeChat Group](https://img.shields.io/badge/Join_WeChat_Group-Community_&_Discussion-07C160?style=for-the-badge&logo=wechat&logoColor=white)](https://zitniklab.hms.harvard.edu/bioagent/wechat_community.html)
87
+
88
+ [![Follow on LinkedIn](https://img.shields.io/badge/Follow_on_LinkedIn-Professional_Updates-0077B5?style=for-the-badge&logo=linkedin&logoColor=white)](https://www.linkedin.com/in/tooluniverse-at-harvard-b9aa88385/)
89
+ [![Follow on X](https://img.shields.io/badge/Follow_on_X-Latest_Updates-000000?style=for-the-badge&logo=x&logoColor=white)](https://x.com/ScientistTools)
90
+
91
+ </div>
92
+
93
+ → **Complete Tool List**: [Available Tools](https://zitniklab.hms.harvard.edu/bioagent/tools/tools_config_index.html)
94
+
95
+ → **Check our website for direct tool usage**: [AIScientist.Tools](https://aiscientist.tools/)
96
+
97
+
98
+ ## 🤖 Building AI Scientists with ToolUniverse in 5 minutes
99
+ - **[Overview](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/index.html)**: Create AI research assistants from any LLM
100
+ - **[Claude Desktop Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/claude_desktop.html)**: Native MCP integration with Claude Desktop App
101
+ - **[Claude Code Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/claude_code.html)**: AI scientist development in Claude Code environment
102
+ - **[Gemini CLI Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/gemini_cli.html)**: Command-line scientific research with Google Gemini
103
+ - **[Qwen Code Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/qwen_code.html)**: AI scientist workflows in Qwen Code environment
104
+ - **[GPT Codex CLI Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/codex_cli.html)**: Terminal-based research with OpenAI Codex
105
+ - **[ChatGPT API Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/chatgpt_api.html)**: Programmatic research with ChatGPT function calling
106
+
107
+
108
+ ## 🔬 What is ToolUniverse?
80
109
 
81
110
  AI scientists are emerging computational systems that serve as collaborative partners in discovery. However, these systems remain difficult to build because they are bespoke, tied to rigid workflows, and lack shared environments that unify tools, data, and analysts into a common ecosystem.
82
111
 
@@ -84,28 +113,32 @@ ToolUniverse addresses this challenge by providing a standardized ecosystem that
84
113
 
85
114
  **Key Features:**
86
115
 
87
- - **AI-Tool Interaction Protocol**: Standardized interface governing how AI scientists issue tool requests and receive results
88
- - **Universal AI Model Support**: Works with any LLM, AI agent, or large reasoning model (GPT, Claude, Gemini, open models)
89
- - **Find Tool & Call Tool Operations**: Maps natural-language descriptions to tool specifications and executes tools with structured results
90
- - **Tool Composition & Workflows**: Chains tools for sequential or parallel execution in self-directed workflows
91
- - **Continuous Expansion**: New tools can be registered locally or remotely without additional configuration
116
+ - [**AI-Tool Interaction Protocol**](https://zitniklab.hms.harvard.edu/bioagent/guide/interaction_protocol.html): Standardized interface governing how AI scientists issue tool requests and receive results
117
+ - [**Universal AI Model Support**](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/index.html): Works with any LLM, AI agent, or large reasoning model (GPT5, Claude, Gemini, Qwen, Deepseek, open models)
118
+ - [**Easy to Load & Find & Call Tool**](https://zitniklab.hms.harvard.edu/bioagent/guide/index.html) (*[WebService](https://aiscientist.tools/), [PythonAPI](https://zitniklab.hms.harvard.edu/bioagent/api/modules.html), [MCP](https://zitniklab.hms.harvard.edu/bioagent/guide/mcp_support.html)*): Maps natural-language descriptions to tool specifications and executes tools with structured results
119
+ - [**Tool Composition & Scientific Workflows**](https://zitniklab.hms.harvard.edu/bioagent/guide/tool_composition.html): Chains tools for sequential or parallel execution in self-directed scientific workflows
120
+ - [**Continuous Expansion**](https://zitniklab.hms.harvard.edu/bioagent/expand_tooluniverse/index.html): New tools can be easily registered locally or remotely without additional configuration
121
+ - [**Multi-Agent Tool Creation & Optimization**](https://zitniklab.hms.harvard.edu/bioagent/guide/scientific_workflows.html): Multi-agent powered tool construction and iterative tool optimization
122
+
123
+
124
+
92
125
 
93
126
  <!-- ![TxAgent](img/TxAgent_ToolUniverse.jpg) -->
94
127
 
95
128
  ## 🚀 Quick Start
96
129
 
97
- **Build your first AI scientist in 5 minutes:**
98
-
99
130
  ```python
100
131
  # 1. Install ToolUniverse
101
132
  pip install tooluniverse
102
133
 
103
- # 2. Create AI scientist system
134
+ # 2. Load tooluniverse
104
135
  from tooluniverse import ToolUniverse
105
136
 
106
137
  tu = ToolUniverse()
107
138
  tu.load_tools() # Load 600+ scientific tools
108
139
 
140
+ # Your AI scientist can reason, experiment, and collaborate in discovery using the AI-tool interaction protocol.
141
+
109
142
  # 3. Use Find Tool operation to discover relevant tools
110
143
  tools = tu.run({
111
144
  "name": "Tool_Finder_Keyword",
@@ -115,36 +148,19 @@ tools = tu.run({
115
148
  # 4. Use Call Tool operation to execute selected tool
116
149
  result = tu.run({
117
150
  "name": "OpenTargets_get_associated_targets_by_disease_efoId",
118
- "arguments": {"efoId": "EFO_0000249"} # Alzheimer's disease
151
+ "arguments": {"efoId": "EFO_0000249"}
119
152
  })
120
153
  ```
121
154
 
122
- **Success!** Your AI scientist can now reason, experiment, and collaborate in discovery using the AI-tool interaction protocol.
123
-
124
- → **Complete Quick Start Tutorial**: [Quick Start Tutorial](https://zitniklab.hms.harvard.edu/bioagent/quickstart.html)
155
+ → **Complete Quick Start Tutorial**: [Quick Start Tutorial](https://zitniklab.hms.harvard.edu/bioagent/quickstart.html) [Getting Started Tutorial](https://zitniklab.hms.harvard.edu/bioagent/getting_started.html)
125
156
 
126
157
 
127
158
  ## 📦 Installation
128
159
 
129
- ### Recommended Installation (uv)
160
+ ### Installation (uv)
130
161
 
131
162
  ```bash
132
- uv add tooluniverse
133
- ```
134
-
135
- ### Standard Installation (pip)
136
-
137
- ```bash
138
- pip install tooluniverse
139
- ```
140
-
141
-
142
- ### Development Installation
143
-
144
- ```bash
145
- git clone https://github.com/mims-harvard/ToolUniverse.git
146
- cd ToolUniverse
147
- uv sync # or pip install -e .[dev]
163
+ uv pip install tooluniverse
148
164
  ```
149
165
 
150
166
  → **Complete Installation Tutorial**: [Installation Tutorial](https://zitniklab.hms.harvard.edu/bioagent/installation.html)
@@ -164,7 +180,7 @@ tu.load_tools()
164
180
 
165
181
  # Find relevant tools
166
182
  tools = tu.run({
167
- "name": "Tool_Finder_Keyword",
183
+ "name": "Tool_Finder_Keyword", # Tool_Finder (Embedding model, GPU required), Tool_Finder_LLM (LLM-based model)
168
184
  "arguments": {"query": "protein structure prediction"}
169
185
  })
170
186
 
@@ -175,38 +191,60 @@ result = tu.run({
175
191
  })
176
192
  ```
177
193
 
178
- ### AI Assistant Integration
194
+ → **Complete Tutorials**: [Installation Tutorial](https://zitniklab.hms.harvard.edu/bioagent/installation.html)
179
195
 
180
- **🖥️ Claude Desktop**: Native integration with Claude AI assistant → [Setup Tutorial](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/claude_desktop.html)
181
- **💻 Claude Code**: Claude Code environment integration → [Setup Tutorial](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/claude_code.html)
182
- **🔮 Gemini CLI**: Google's AI agent integration → [Setup Tutorial](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/gemini_cli.html)
183
- **🧠 Qwen Code**: AI scientist integration with Qwen Code environment → [Setup Tutorial](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/qwen_code.html)
184
- **⚡ GPT Codex CLI**: Terminal-based AI scientist with Codex CLI → [Setup Tutorial](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/codex_cli.html)
185
- **🎯 ChatGPT API**: OpenAI integration via MCP protocol and function calling → [Setup Tutorial](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/chatgpt_api.html)
196
+ ### MCP Support
186
197
 
187
- → **AI Assistant Setup Guides**: [Building AI Scientists](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/index.html)
198
+ ```python
199
+ # run one command to launch the tooluniverse mcp server
200
+ tooluniverse-smcp
201
+ ```
202
+ → **Complete Tutorial**: [MCP Support](https://zitniklab.hms.harvard.edu/bioagent/guide/mcp_support.html)
188
203
 
189
- ### MCP Server Integration
190
204
 
191
- ```bash
192
- # Start MCP server
193
- python -m tooluniverse.smcp_server
205
+ ## 🚀 AI Scientists Projects Powered by ToolUniverse
194
206
 
195
- # Or with CLI
196
- tooluniverse-mcp --help
197
- ```
207
+ *Building your own project with ToolUniverse? We'd love to feature it here! Submit your project via [GitHub Pull Request](https://github.com/mims-harvard/ToolUniverse/pulls) or contact us.*
208
+
209
+ ---
210
+ **TxAgent: AI Agent for Therapeutic Reasoning** [[Project]](https://zitniklab.hms.harvard.edu/TxAgent) [[Paper]](https://arxiv.org/pdf/2503.10970) [[PiPy]](https://pypi.org/project/txagent/) [[Github]](https://github.com/mims-harvard/TxAgent) [[HuggingFace]](https://huggingface.co/collections/mims-harvard/txagent-67c8e54a9d03a429bb0c622c)
211
+ > **TxAgent** is an AI agent for therapeutic reasoning that leverages ToolUniverse's comprehensive scientific tool ecosystem to solve complex therapeutic reasoning tasks.
198
212
 
199
- ### Core Capabilities
200
213
 
201
- - **🔍 Find Tool Operation**: Maps natural-language descriptions to tool specifications using semantic search
202
- - **⚙️ Call Tool Operation**: Executes selected tools with structured arguments and returns text, embeddings, or JSON
203
- - **🔗 Tool Composition**: Chains tools for sequential or parallel execution in self-directed workflows
204
- - **🛠️ Tool Creation**: Automatically creates new tools from natural language descriptions
205
- - **🔄 Iterative Optimization**: Refines tool interfaces and specifications for correct use by AI scientists
206
- - **🌐 Shared Environment**: Unifies tools, data, and analysts into a common ecosystem for interoperability and reuse
214
+ ---
215
+ **Hypercholesterolemia Drug Discovery** [[Tutorial]]((https://zitniklab.hms.harvard.edu/bioagent/tutorials/tooluniverse_case_study.html)) [[Code]](https://colab.research.google.com/drive/1UwJ6RwyUoqI5risKQ365EeFdDQWOeOCv?usp=sharing)
216
+
217
+ ---
218
+
219
+
220
+ ## 🤝 Contribution and Community
221
+
222
+ **We are actively looking for core contributors for ToolUniverse!**
223
+ Please join our [Slack Channel](https://join.slack.com/t/tooluniversehq/shared_invite/zt-3dic3eoio-5xxoJch7TLNibNQn5_AREQ) or reach out to [Shanghua Gao](mailto:shanghuagao@gmail.com)/[Marinka Zitnik](mailto:marinka@hms.harvard.edu).
224
+
225
+ **Get Involved:**
226
+
227
+ - **Report Issues**: [GitHub Issues](https://github.com/mims-harvard/ToolUniverse/issues)
228
+ - **Join Discussions**: [Slack Channel](https://github.com/mims-harvard/ToolUniverse/discussions)
229
+ - **Contact**: Reach out to [Shanghua Gao](shanghuagao@gmail.com)/[Marinka Zitnik](marinka@hms.harvard.edu)
230
+ - **Contribute**: See our [Contributing Tutorial](https://zitniklab.hms.harvard.edu/bioagent/expand_tooluniverse/comprehensive_tool_guide.html)
231
+
232
+
233
+
234
+ ### Contributors
235
+
236
+ - **[Shanghua Gao](https://shgao.site)**
237
+ - **[Richard Zhu](https://www.linkedin.com/in/richard-zhu-4236901a7/)**
238
+ - **[Pengwei Sui](mailto:pengwei_sui@hms.harvard.edu)**
239
+ - **[Zhenglun Kong](https://zlkong.github.io/homepage/)**
240
+ - **[Sufian Aldogom](mailto:saldogom@mit.edu)**
241
+ - **[Yepeng Huang](mailto:yepeng_huang@hms.harvard.edu)**
242
+ - **[Ayush Noori](https://www.ayushnoori.com/)**
243
+ - **[Reza Shamji](mailto:reza_shamji@hms.harvard.edu)**
244
+ - **[Krishna Parvataneni](mailto:krishna_parvataneni@hms.harvard.edu)**
245
+ - **[Theodoros Tsiligkaridis](https://sites.google.com/view/theo-t)**
246
+ - **[Marinka Zitnik](https://zitniklab.hms.harvard.edu/)**
207
247
 
208
- → **Detailed Usage Tutorial**: [README_USAGE.md](README_USAGE.md)
209
- → **Complete Getting Started**: [Getting Started Tutorial](https://zitniklab.hms.harvard.edu/bioagent/getting_started.html)
210
248
 
211
249
 
212
250
  ## 📚 Documentation
@@ -225,108 +263,39 @@ Our comprehensive documentation covers everything from quick start to advanced w
225
263
  - **[Tool Caller](https://zitniklab.hms.harvard.edu/bioagent/guide/tool_caller.html)**: Primary execution engine
226
264
  - **[Tool Composition](https://zitniklab.hms.harvard.edu/bioagent/guide/tool_composition.html)**: Chain tools into workflows
227
265
  - **[Scientific Workflows](https://zitniklab.hms.harvard.edu/bioagent/guide/scientific_workflows.html)**: Real-world research scenarios
228
-
229
- ### 🤖 Building AI Scientists
230
- - **[Overview](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/index.html)**: Create AI research assistants from any LLM
231
- - **[Claude Desktop Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/claude_desktop.html)**: Native MCP integration with Claude Desktop App
232
- - **[Claude Code Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/claude_code.html)**: AI scientist development in Claude Code environment
233
- - **[Gemini CLI Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/gemini_cli.html)**: Command-line scientific research with Google Gemini
234
- - **[Qwen Code Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/qwen_code.html)**: AI scientist workflows in Qwen Code environment
235
- - **[GPT Codex CLI Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/codex_cli.html)**: Terminal-based research with OpenAI Codex
236
- - **[ChatGPT API Integration](https://zitniklab.hms.harvard.edu/bioagent/guide/building_ai_scientists/chatgpt_api.html)**: Programmatic research with ChatGPT function calling
266
+ - **[MCP Support](https://zitniklab.hms.harvard.edu/bioagent/guide/mcp_support.html)**: Model Context Protocol integration
267
+ - **[Logging](https://zitniklab.hms.harvard.edu/bioagent/guide/logging.html)**: Comprehensive logging configuration
237
268
 
238
269
  ### 🛠️ Advanced Features
239
270
  - **[Hooks System](https://zitniklab.hms.harvard.edu/bioagent/guide/hooks/index.html)**: Intelligent output processing
271
+ - **[Hook Configuration](https://zitniklab.hms.harvard.edu/bioagent/guide/hooks/hook_configuration.html)**: Configure hooks for different outputs
272
+ - **[File Save Hook](https://zitniklab.hms.harvard.edu/bioagent/guide/hooks/file_save_hook.html)**: Save tool outputs to files
273
+ - **[Summarization Hook](https://zitniklab.hms.harvard.edu/bioagent/guide/hooks/summarization_hook.html)**: Summarize tool outputs
274
+ - **[Server Stdio Hooks](https://zitniklab.hms.harvard.edu/bioagent/guide/hooks/server_stdio_hooks.html)**: Server communication hooks
240
275
  - **[Expert Feedback](https://zitniklab.hms.harvard.edu/bioagent/tutorials/expert_feedback.html)**: Human-in-the-loop consultation
241
276
  - **[Agentic Tools](https://zitniklab.hms.harvard.edu/bioagent/tutorials/agentic_tools_tutorial.html)**: AI-powered tool development
242
277
  - **[Case Study](https://zitniklab.hms.harvard.edu/bioagent/tutorials/tooluniverse_case_study.html)**: End-to-end drug discovery workflow
278
+ - **[Tool Optimization](https://zitniklab.hms.harvard.edu/bioagent/tutorials/optimization/Tool_Description_Optimizer_Tutorial.html)**: Optimize tool descriptions for better performance
243
279
 
244
280
  ### 🔧 Expanding ToolUniverse
245
281
  - **[Architecture](https://zitniklab.hms.harvard.edu/bioagent/expand_tooluniverse/architecture.html)**: System architecture overview
246
282
  - **[Local Tool Registration](https://zitniklab.hms.harvard.edu/bioagent/expand_tooluniverse/local_tool_registration.html)**: Create custom tools
247
283
  - **[Remote Tool Registration](https://zitniklab.hms.harvard.edu/bioagent/expand_tooluniverse/remote_tool_registration.html)**: Integrate external services
248
- - **[Contributing Tools](https://zitniklab.hms.harvard.edu/bioagent/expand_tooluniverse/comprehensive_tool_guide.html)**: Complete contribution Tutorial
284
+ - **[Contributing Tools](https://zitniklab.hms.harvard.edu/bioagent/expand_tooluniverse/comprehensive_tool_guide.html)**: Complete contribution guide
285
+ - **[Adding Tools Tutorial](https://zitniklab.hms.harvard.edu/bioagent/tutorials/addtools/Adding_Tools_Tutorial.html)**: Step-by-step tool addition guide
286
+ - **[MCP Tool Registration](https://zitniklab.hms.harvard.edu/bioagent/tutorials/addtools/mcp_tool_registration_en.html)**: Register tools via MCP
249
287
 
250
288
  ### 📚 API Reference
251
- - **[API Directory](https://zitniklab.hms.harvard.edu/bioagent/api_directory.html)**: Complete tool listing
252
- - **[Tools Reference](https://zitniklab.hms.harvard.edu/bioagent/api/tools_complete_reference.html)**: Detailed API documentation
253
- - **[Quick Reference](https://zitniklab.hms.harvard.edu/bioagent/api_quick_reference.html)**: Essential API Tutorial
254
-
255
- ### 🆘 Help & Support
256
- - **[FAQ](https://zitniklab.hms.harvard.edu/bioagent/faq.html)**: Frequently asked questions
257
- - **[Troubleshooting](https://zitniklab.hms.harvard.edu/bioagent/troubleshooting.html)**: Common issues and solutions
258
- - **[Help Index](https://zitniklab.hms.harvard.edu/bioagent/help/index.html)**: Support resources
289
+ - **[API Directory](https://zitniklab.hms.harvard.edu/bioagent/api/modules.html)**: Complete module listing
290
+ - **[Core Modules](https://zitniklab.hms.harvard.edu/bioagent/api/tooluniverse.html)**: Main ToolUniverse class and utilities
291
+ - **[Tool Classes](https://zitniklab.hms.harvard.edu/bioagent/api/tooluniverse.base_tool.html)**: Base tool classes and implementations
292
+ - **[Compose Scripts](https://zitniklab.hms.harvard.edu/bioagent/api/tooluniverse.compose_scripts.html)**: Tool composition utilities
293
+ - **[MCP Integration](https://zitniklab.hms.harvard.edu/bioagent/api/tooluniverse.mcp_integration.html)**: Model Context Protocol support
259
294
 
260
295
  → **Browse All Documentation**: [ToolUniverse Documentation](https://zitniklab.hms.harvard.edu/bioagent/)
261
296
 
262
- ## 🚀 AI Scientists Projects Powered by ToolUniverse
263
-
264
- ### TxAgent: AI Agent for Therapeutic Reasoning
265
-
266
- **TxAgent** is an AI agent for therapeutic reasoning that leverages ToolUniverse's comprehensive scientific tool ecosystem to solve complex therapeutic reasoning tasks.
267
-
268
- [![ProjectPage](https://img.shields.io/badge/Page-TxAgent-red)](https://zitniklab.hms.harvard.edu/TxAgent) [![PaperLink](https://img.shields.io/badge/Arxiv-TxAgent-red)](https://arxiv.org/pdf/2503.10970) [![TxAgent-PIP](https://img.shields.io/badge/Pip-TxAgent-blue)](https://pypi.org/project/txagent/) [![TxAgent](https://img.shields.io/badge/Code-TxAgent-purple)](https://github.com/mims-harvard/TxAgent) [![HuggingFace](https://img.shields.io/badge/HuggingFace-TxAgentT1-yellow)](https://huggingface.co/collections/mims-harvard/txagent-67c8e54a9d03a429bb0c622c)
269
-
270
- ---
271
-
272
- *Building your own project with ToolUniverse? We'd love to feature it here! Submit your project via [GitHub Issues](https://github.com/mims-harvard/ToolUniverse/issues) or contact us.*
273
-
274
- ## 🌍 Ecosystem & Community
275
-
276
- ### Democratizing AI Agents for Science
277
-
278
- ToolUniverse addresses the fundamental challenge that AI scientists remain difficult to build because they are bespoke, tied to rigid workflows, and lack shared environments. Drawing inspiration from how unified ecosystems have transformed omics research by enabling interoperability, reuse, and community-driven development, ToolUniverse provides comparable infrastructure for AI scientists.
279
-
280
- **Core Principles:**
281
-
282
- - **🔓 Unified Ecosystem**: Integrates tools, data, and analysts into a common environment like omics platforms
283
- - **📐 Standardized Protocol**: AI-tool interaction protocol analogous to HTTP for internet communication
284
- - **🔗 Interoperability**: Tool specification schema that standardizes definitions and enables consistent inference
285
- - **🌐 Collaborative Discovery**: Enables AI scientists to serve as computational partners in research
286
- - **🚀 Scalable Construction**: Builds AI scientists at scale without bespoke implementation
287
297
 
288
- ### Research Applications
289
-
290
- **🧬 Drug Discovery & Development**
291
- - Target identification and validation
292
- - Compound screening and optimization
293
- - Safety and toxicity assessment
294
- - Clinical trial analysis and outcomes
295
- - **Case Study**: Hypercholesterolemia research where ToolUniverse enabled an AI scientist to identify a potent drug analog with favorable predicted properties
296
-
297
- **🔬 Molecular Biology Research**
298
- - Gene function annotation and analysis
299
- - Protein structure prediction and modeling
300
- - Pathway analysis and network construction
301
- - Disease association studies
302
-
303
- **📊 Literature & Knowledge Mining**
304
- - Automated systematic literature reviews
305
- - Named entity recognition and extraction
306
- - Citation network analysis and mapping
307
- - Patent landscape analysis and IP research
308
-
309
- **🤖 AI-Driven Research Automation**
310
- - Hypothesis generation and testing
311
- - Experimental design optimization
312
- - Multi-modal data analysis and integration
313
- - Results interpretation and reporting
314
-
315
- ### Community & Support
316
-
317
- **Get Involved:**
318
- - 🐛 **Report Issues**: [GitHub Issues](https://github.com/mims-harvard/ToolUniverse/issues)
319
- - 💬 **Join Discussions**: [GitHub Discussions](https://github.com/mims-harvard/ToolUniverse/discussions)
320
- - 📧 **Contact**: Visit [aiscientist.tools](https://aiscientist.tools) for more information
321
- - 🤝 **Contribute**: See our [Contributing Tutorial](https://zitniklab.hms.harvard.edu/bioagent/expand_tooluniverse/comprehensive_tool_guide.html)
322
-
323
- **Integrations & Extensions:**
324
- - **Model Context Protocol (MCP)**: Full MCP server support for AI assistant integration
325
- - **Claude Desktop**: Native integration with Anthropic's Claude Desktop App
326
- - **OpenAI Function Calling**: Direct integration with ChatGPT and GPT models
327
- - **Custom Agents**: Build domain-specific AI research assistants
328
-
329
- ## Citation
298
+ ### Citation
330
299
 
331
300
  ```
332
301
  @misc{gao2025txagent,
@@ -340,31 +309,6 @@ ToolUniverse addresses the fundamental challenge that AI scientists remain diffi
340
309
  }
341
310
  ```
342
311
 
343
- ## Contact & Contributors
344
-
345
- ### Contact
346
-
347
- For questions, suggestions, or collaboration opportunities:
348
-
349
- - **📧 Contact**: [Shanghua Gao](mailto:shanghuagao@gmail.com)
350
- - **🌐 Web Platform**: [aiscientist.tools](https://aiscientist.tools)
351
- - **💻 GitHub**: [github.com/mims-harvard/ToolUniverse](https://github.com/mims-harvard/ToolUniverse)
352
-
353
- ### Contributors
354
-
355
- - **[Shanghua Gao](https://shgao.site)**
356
- - **[Richard Zhu](https://www.linkedin.com/in/richard-zhu-4236901a7/)**
357
- - **[Pengwei Sui](mailto:pengwei_sui@hms.harvard.edu)**
358
- - **[Zhenglun Kong](https://zlkong.github.io/homepage/)**
359
- - **[Sufian Aldogom](mailto:saldogom@mit.edu)**
360
- - **[Yepeng Huang](mailto:yepeng_huang@hms.harvard.edu)**
361
- - **[Ayush Noori](https://www.ayushnoori.com/)**
362
- - **[Reza Shamji](mailto:reza_shamji@hms.harvard.edu)**
363
- - **[Krishna Parvataneni](mailto:krishna_parvataneni@hms.harvard.edu)**
364
- - **[Theodoros Tsiligkaridis](https://sites.google.com/view/theo-t)**
365
- - **[Marinka Zitnik](https://zitniklab.hms.harvard.edu/)**
366
-
367
-
368
312
 
369
313
  ### Acknowledgments
370
314
 
@@ -372,6 +316,4 @@ ToolUniverse is developed by the [Zitnik Lab](https://zitniklab.hms.harvard.edu/
372
316
 
373
317
  ---
374
318
 
375
- **🧬🤖🔬 Happy researching with ToolUniverse!**
376
-
377
- *Democratizing AI agents for science - one tool at a time.*
319
+ *Democratizing AI agents for science - one tooluniverse at a time.*
{tooluniverse-1.0.0.dist-info → tooluniverse-1.0.2.dist-info}/RECORD RENAMED
@@ -1,7 +1,7 @@
1
- tooluniverse/__init__.py,sha256=uQtjC0aSwvfVxR_dxbvptPyoJ2BUQ1_kIm39dnJkUiU,13175
1
+ tooluniverse/__init__.py,sha256=DXdZsB6gSGAubWpPjZDJZGo_PeH8l3BeJ-NeKsuY5Co,13944
2
2
  tooluniverse/admetai_tool.py,sha256=nEfq7dwvFqjDryKv3nivGoBSRap6LT7_m76iMu72Q6E,3009
3
3
  tooluniverse/agentic_tool.py,sha256=6qMPrcwVAjSobgmeuez3Dy6ZMgJCIkVrBdcMPuE44cU,23166
4
- tooluniverse/alphafold_tool.py,sha256=xvtFvj52UDXdnV9oPRj0XJLThLWyLshjmebQIqnWOq8,3467
4
+ tooluniverse/alphafold_tool.py,sha256=NkKQttC0MfuetCTTcWTFEG2MdhyKwy3jb5TR4odiqAw,4658
5
5
  tooluniverse/base_tool.py,sha256=uA-wGrs1drEJJbIp24YfWWWWfNGeMqEQo1tI6r3csEg,5664
6
6
  tooluniverse/boltz_tool.py,sha256=Kfcymeu6FRRpWLw7dWhEON4PUBTsHuBiTAPJlYH5ZGs,7930
7
7
  tooluniverse/chem_tool.py,sha256=tJusjamWNp0x8Ke-cvVgKB7-clvD6LXTaXA0Q0fqQv4,8336
@@ -61,15 +61,17 @@ tooluniverse/compose_scripts/output_summarizer.py,sha256=P6wS5dRePJfWaU5Rh9S-uPe
61
61
  tooluniverse/compose_scripts/tool_description_optimizer.py,sha256=L6Kd8e6Af8pD6dy8jrFzJq5jq-QNbg72gkQktn8IDBE,32175
62
62
  tooluniverse/compose_scripts/tool_discover.py,sha256=RKzCmhDWzDCsAl2KCcEQRNH96mOf_2bTQOH2EVGvpC8,24976
63
63
  tooluniverse/compose_scripts/tool_graph_composer.py,sha256=lmGeN94Mc9SZCeYo8ez55tKYLS1-dI3JDm-onKB5DZA,15566
64
+ tooluniverse/compose_scripts/tool_graph_generation.py,sha256=zPVN58GufflVCo8Ak3V9_DpNjS9oLrhQ5A_JCN9TCzY,10052
65
+ tooluniverse/compose_scripts/tool_metadata_generator.py,sha256=GCd_zKxN5n7vz_5Zlj7nP2KpGzfi0baRfOnKt9ZQsRI,19004
64
66
  tooluniverse/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
65
67
  tooluniverse/data/admetai_tools.json,sha256=08LCNSCZJxL6ZuPF1mVfbdlh9f2X5tnEwIYU7RcLLSM,6497
66
- tooluniverse/data/agentic_tools.json,sha256=er40nXt19HS4S3N6jbpRmTPhprBbAsWwRB2lhngFsAI,89865
67
- tooluniverse/data/alphafold_tools.json,sha256=1q0nB2DXyAPQxLoRXWnkFKYBgLW9AGAGMP7qqOuOx7A,2678
68
+ tooluniverse/data/adverse_event_tools.json,sha256=EDU_X8r4YtMnVt7SJpyimdo3FlcSNFSdvckWLShDODA,12752
69
+ tooluniverse/data/agentic_tools.json,sha256=2LVwit3QqoHQRja2f_8KvRh9evyIV62fpYqo4WyiIeM,103309
70
+ tooluniverse/data/alphafold_tools.json,sha256=--JO9z9fQM95t5dphrBoKKr28HKWg7iQOUm1hQYhEOk,10628
68
71
  tooluniverse/data/boltz_tools.json,sha256=_pMSRBkO3uj4vjYnQ7H6qbZFa4rzc9Looj6zHRyIkMg,303
69
72
  tooluniverse/data/chembl_tools.json,sha256=aQfFyFlUNz_y9ZTO3dNje4QGhqt7rLv6j5YdMJR2-x8,747
70
- tooluniverse/data/clait_tools.json,sha256=EDU_X8r4YtMnVt7SJpyimdo3FlcSNFSdvckWLShDODA,12752
71
73
  tooluniverse/data/clinicaltrials_gov_tools.json,sha256=_3NphPP_9cLviktCNnfsoSo_EHIkKA4lDTmuSycQJ58,16102
72
- tooluniverse/data/compose_tools.json,sha256=LyiugKvAzfCMs02ZkY7EHbk92UzaSdDcFocaGuVS8xU,7793
74
+ tooluniverse/data/compose_tools.json,sha256=9l4FbSx8s0V6-k_6-5Fr2-Z_LDZ_WnxZke_UZxZk1g8,10913
73
75
  tooluniverse/data/dailymed_tools.json,sha256=F7zSeK6MM9eGKUXZpyeVNFD1UkOkd-0OoxmTlgnEffg,2059
74
76
  tooluniverse/data/dataset_tools.json,sha256=MiJKlKTEFRILbhy2DPnhG_9upFu5GJ2-YCu9HarjcD8,29506
75
77
  tooluniverse/data/disease_target_score_tools.json,sha256=xhU5bigeNa2FSn0bqVPArXVid6lkwfFBDpD2xbSGisc,23199
@@ -99,7 +101,7 @@ tooluniverse/data/pubtator_tools.json,sha256=WkcP3DFlC6EcL9zf2KMrDITC0_fYPNTu60b
99
101
  tooluniverse/data/rcsb_pdb_tools.json,sha256=gKY8Y9QDN4-3ILc2KFLGdPFXG1M9LhXMAPFQtwnP7Bg,47767
100
102
  tooluniverse/data/reactome_tools.json,sha256=h8Ubkllr4LMFc8UPUFQ8SZf_-wdWdh_NaHAKhx31Db0,557
101
103
  tooluniverse/data/semantic_scholar_tools.json,sha256=e-gkWftA3MH4ep_WjT2tCncrsSX_x_Y9WtteowMkqRA,1116
102
- tooluniverse/data/special_tools.json,sha256=2QSWknGvpB-t5RIv8NrhiveovxgN-PDmrx-Uea83o9o,937
104
+ tooluniverse/data/special_tools.json,sha256=frBbn-I0pFeTQpJyXqLIGvsu12gIRwasXKZHLSmE2qA,991
103
105
  tooluniverse/data/tool_composition_tools.json,sha256=kx9S7TSQqkQb4n13WuW204NL9Zqi8csA1Jmf_wSQ3p4,6323
104
106
  tooluniverse/data/toolfinderkeyword_defaults.json,sha256=bydTVd0IUSDiilNQ6kZU-cn4lPqOfInfUAxhh8yMXqY,1050
105
107
  tooluniverse/data/txagent_client_tools.json,sha256=GdQk5H6uP_Uj9Ejz3nMdzNhZJV42t75aiRW8oWtkzRw,409
@@ -120,7 +122,6 @@ tooluniverse/data/packages/original_tools.txt,sha256=47DEQpj8HBSa-_TImW-5JCeuQeR
120
122
  tooluniverse/data/packages/physics_astronomy_tools.json,sha256=n0ovkqOXhUcDBUPDGNKrCs3NbYJOcnOEADXCafgI4SE,1527
121
123
  tooluniverse/data/packages/scientific_computing_tools.json,sha256=YhGZIvllzmesF2Pm_OnIY6EVgVJ36nqvQN8Pvno-5BY,28300
122
124
  tooluniverse/data/packages/single_cell_tools.json,sha256=HxHygdgtQM6kObnPdCqUTSwNI8Cux1LoOeYXuxxjjtw,36069
123
- tooluniverse/data/packages/software_tools.json,sha256=y9wb3dts-aQW7RxBDL7EHLSl2vNBuw5LW8nL7VZ6IFU,460958
124
125
  tooluniverse/data/packages/structural_biology_tools.json,sha256=tQw78Ot6i9PWGUX3yVsZOgeOqZOhqTeh9mQVahZaUec,31755
125
126
  tooluniverse/data/packages/visualization_tools.json,sha256=yS-M_H3px6Ea-De-Km8v_gVMPa2Vdt6LmAPU-Zsai3c,25177
126
127
  tooluniverse/remote/boltz/boltz_mcp_server.py,sha256=oAqO8j52M4ntn1Ihdjclj6nJwHxRTbyxIF6efi_G_jU,2088
@@ -129,9 +130,6 @@ tooluniverse/remote/expert_feedback/human_expert_mcp_tools.py,sha256=WeXuIjus7bs
129
130
  tooluniverse/remote/expert_feedback/simple_test.py,sha256=rv_UN4oHH6BNmYVuPKRfPdIEjtdFvG0MXYoNOIHqXvg,535
130
131
  tooluniverse/remote/expert_feedback/start_web_interface.py,sha256=_As-c4HRx9nBb_xjiO040HMj7v8RwIB2f9yT1W4f0Yo,6136
131
132
  tooluniverse/remote/expert_feedback/web_only_interface.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
132
- tooluniverse/remote/expert_feedback_mcp/human_expert_mcp_server.py,sha256=hsI7npDj142_YlLjPfCWE3RYzgrB6hCx_L26StoiDTE,59973
133
- tooluniverse/remote/expert_feedback_mcp/simple_test.py,sha256=tBLjIUhULLBB_sKjf_2hsm9IQkMjcI4Tf4p34ZUen_w,913
134
- tooluniverse/remote/expert_feedback_mcp/start_web_interface.py,sha256=OYnEhU_sHQlUKKxF6HJoJbn220HFws7eS47QVDF9ESg,2738
135
133
  tooluniverse/remote/immune_compass/compass_tool.py,sha256=jpxfMNDH7UWF4FiFzLB-JyfQYNUB8ZA6somvqkEtDH0,14423
136
134
  tooluniverse/remote/pinnacle/pinnacle_tool.py,sha256=Oprd2RyMCZF-NsMjkZMpPOB2apCfws7C3IWIn_U3JxQ,14632
137
135
  tooluniverse/remote/transcriptformer/transcriptformer_tool.py,sha256=ZU_TT340LBwjTuowTaQV6QcANcsjYimA4CVSJHdU0c0,25229
@@ -142,7 +140,7 @@ tooluniverse/scripts/visualize_tool_graph.py,sha256=4USGnUVdffJfOxourBKELldGhhk5
142
140
  tooluniverse/test/mcp_server_test.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
143
141
  tooluniverse/test/test_admetai_tool.py,sha256=Z_J-LjPgvqsdWpHvXwq24rt2e3Pg3IIp6t4XWUmBPoA,31991
144
142
  tooluniverse/test/test_agentic_tool.py,sha256=vMU6uxU0rB6DaYL105XY4euU9VOWWXR3c-qGzMzhHkU,5803
145
- tooluniverse/test/test_alphafold_tool.py,sha256=BSlAyXbnGS8eCocSeQ58KoCYKbj6mADPChxjrwrl-Q4,2236
143
+ tooluniverse/test/test_alphafold_tool.py,sha256=Hj-ITZqWO-U8OVNveVtmLVACiOAzwUYh1COwjcIaxrQ,3459
146
144
  tooluniverse/test/test_chem_tool.py,sha256=N_ARBvouWyJMgWL1yMXSg1w407fBB-CZxJysFVRE6uY,1589
147
145
  tooluniverse/test/test_compose_lieraturereview.py,sha256=qNkgXHXJm7Ekiw9e0mcgtqs1z5c-MeE3qy0RoexF9Tg,1956
148
146
  tooluniverse/test/test_compose_tool.py,sha256=reRhUWhSuWwuorWF0jbDQHhIEd14UTq_GDSe0Wwl0v4,16607
@@ -178,9 +176,9 @@ tooluniverse/test/test_tools_find.py,sha256=vP9EV04iXXmejoY2rDhEM0Yc9uHEO2qEmd8s
178
176
  tooluniverse/test/test_uniprot_tools.py,sha256=V0x7aQjjlE-wS5alFWviHNwC3saHiEE5pMMqAcbokaw,2582
179
177
  tooluniverse/test/test_uspto_tool.py,sha256=y2VQBMbiHCTAmvXhBosHrc1innstUl9UMWNoPaBv-RY,2311
180
178
  tooluniverse/test/test_xml_tool.py,sha256=tx-w_psZ801C2kM6aDbBzR8QV3Vgu1p2zP6zExhjpVs,3602
181
- tooluniverse-1.0.0.dist-info/licenses/LICENSE,sha256=Unq9Y3buGL9J0vFop03FYZiVweytt1xdgLIh2n9ck2c,1115
182
- tooluniverse-1.0.0.dist-info/METADATA,sha256=Np2joe8Nr0P2OclPRLMuoxZsR4G2eKnnX6S5XHE-Hfs,19452
183
- tooluniverse-1.0.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
184
- tooluniverse-1.0.0.dist-info/entry_points.txt,sha256=Z_znqzUc_XR9ZIXViavqNAwBkwM4XYfC9h7LoBbAAog,600
185
- tooluniverse-1.0.0.dist-info/top_level.txt,sha256=zZ8YeCJ5FAkEwdd_mxsFtSCQMBDgBdxrrmHo3RNBiWs,13
186
- tooluniverse-1.0.0.dist-info/RECORD,,
179
+ tooluniverse-1.0.2.dist-info/licenses/LICENSE,sha256=Unq9Y3buGL9J0vFop03FYZiVweytt1xdgLIh2n9ck2c,1115
180
+ tooluniverse-1.0.2.dist-info/METADATA,sha256=SEmSfNFe6gFqwsmk5IyQbtAF2EVknHL2nm7Y7McWUUM,18270
181
+ tooluniverse-1.0.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
182
+ tooluniverse-1.0.2.dist-info/entry_points.txt,sha256=Z_znqzUc_XR9ZIXViavqNAwBkwM4XYfC9h7LoBbAAog,600
183
+ tooluniverse-1.0.2.dist-info/top_level.txt,sha256=zZ8YeCJ5FAkEwdd_mxsFtSCQMBDgBdxrrmHo3RNBiWs,13
184
+ tooluniverse-1.0.2.dist-info/RECORD,,