|
|
--- |
|
|
license: mit |
|
|
task_categories: |
|
|
- feature-extraction |
|
|
- zero-shot-classification |
|
|
language: |
|
|
- en |
|
|
tags: |
|
|
- agent |
|
|
- tools |
|
|
- function-calling |
|
|
- api |
|
|
size_categories: |
|
|
- 1K<n<10K |
|
|
--- |
|
|
|
|
|
# SAGE Agent Tools Dataset |
|
|
|
|
|
Curated catalog of 1200+ agent tools with structured metadata for tool selection and usage. |
|
|
|
|
|
## Dataset Description |
|
|
|
|
|
This dataset provides a comprehensive catalog of tools that AI agents can use, including APIs, utilities, and services across multiple categories. |
|
|
|
|
|
## Usage |
|
|
|
|
|
```python |
|
|
from datasets import load_dataset |
|
|
|
|
|
dataset = load_dataset("intellistream/sage-agent-tools") |
|
|
|
|
|
# Access tools |
|
|
for tool in dataset['train']: |
|
|
print(tool['tool_id'], tool['tool_name']) |
|
|
print(tool['category']) |
|
|
``` |
|
|
|
|
|
## Fields |
|
|
|
|
|
- `tool_id`: Unique identifier |
|
|
- `tool_name`: Human-readable name |
|
|
- `description`: Tool functionality description |
|
|
- `category`: Tool category (e.g., "weather", "search") |
|
|
- `capabilities`: List of capabilities (JSON string) |
|
|
- `parameters`: JSON schema for parameters (JSON string) |
|
|
- `api_endpoint`: API URL (if applicable) |
|
|
|
|
|
## Citation |
|
|
|
|
|
```bibtex |
|
|
@software{sage_data_2026, |
|
|
author = {IntelliStream Team}, |
|
|
title = {SAGE Data: Unified Dataset Management}, |
|
|
year = {2026}, |
|
|
url = {https://github.com/intellistream/sageData} |
|
|
} |
|
|
``` |
|
|
|
