Skill: Funding Trend Forecaster
ID: 200
Version: 1.0.0
Author: OpenClaw Agent
License: MIT
Overview
Funding Trend Forecaster is an intelligent analysis tool that uses Natural Language Processing (NLP) technology to analyze awarded project abstracts from major global research funding agencies (NIH, NSF, Horizon Europe) and predict funding preference shift trends for the next 3-5 years.
Features
- - Multi-source Data Collection: Automatically fetches awarded project data from NIH, NSF, Horizon Europe
- NLP Deep Analysis: Uses advanced text mining techniques to extract topics, keywords, and research trends
- Trend Prediction Model: Predicts funding direction changes based on time series analysis and topic modeling
- Visualized Reports: Generates charts and trend reports for intuitive display of analysis results
- Field Segmentation: Categorized analysis by medicine, engineering, natural sciences, and other fields
Installation
CODEBLOCK0
Dependencies
CODEBLOCK1
Usage
Command Line Interface
CODEBLOCK2
API Call
CODEBLOCK3
Parameters
| Parameter | Type | Default | Required | Description |
|---|
| INLINECODE0 | flag | false | No | Run full analysis workflow on all sources |
| INLINECODE1 |
string | - | No | Specific agency to analyze (nih, nsf, horizon_europe) |
|
--months | int | 6 | No | Number of months of historical data to analyze |
|
--years | int | 5 | No | Years ahead for trend prediction |
|
--visualize | flag | false | No | Generate visualization charts |
|
--forecast | flag | false | No | Generate trend forecast |
|
--input,
-i | string | - | No | Input data file path (for visualization/forecast) |
|
--output,
-o | string | - | No | Output file path |
|
--config | string | config.json | No | Path to configuration file |
Data Sources
| Agency | Data Source URL | Update Frequency |
|---|
| NIH | https://reporter.nih.gov/ | Daily |
| NSF |
https://www.nsf.gov/awardsearch/ | Daily |
| Horizon Europe | https://ec.europa.eu/info/funding-tenders/opportunities/ | Weekly |
Configuration
Create config.json file to customize analysis parameters:
CODEBLOCK4
Output Format
JSON Report Structure
CODEBLOCK5
Architecture
CODEBLOCK6
Roadmap
- - [x] Basic architecture design
- [x] Core analysis module
- [ ] More data source support (Wellcome Trust, JSPS, etc.)
- [ ] Real-time data stream processing
- [ ] Interactive web interface
- [ ] Machine learning model optimization
License
MIT License - See LICENSE file in project root directory
Generated by OpenClaw Agent | Skill ID: 200
Risk Assessment
| Risk Indicator | Assessment | Level |
|---|
| Code Execution | Python scripts with tools | High |
| Network Access |
External API calls | High |
| File System Access | Read/write data | Medium |
| Instruction Tampering | Standard prompt guidelines | Low |
| Data Exposure | Data handled securely | Medium |
Security Checklist
- - [ ] No hardcoded credentials or API keys
- [ ] No unauthorized file system access (../)
- [ ] Output does not expose sensitive information
- [ ] Prompt injection protections in place
- [ ] API requests use HTTPS only
- [ ] Input validated against allowed patterns
- [ ] API timeout and retry mechanisms implemented
- [ ] Output directory restricted to workspace
- [ ] Script execution in sandboxed environment
- [ ] Error messages sanitized (no internal paths exposed)
- [ ] Dependencies audited
- [ ] No exposure of internal service architecture
Prerequisites
CODEBLOCK7
Evaluation Criteria
Success Metrics
- - [ ] Successfully executes main functionality
- [ ] Output meets quality standards
- [ ] Handles edge cases gracefully
- [ ] Performance is acceptable
Test Cases
- 1. Basic Functionality: Standard input → Expected output
- Edge Case: Invalid input → Graceful error handling
- Performance: Large dataset → Acceptable processing time
Lifecycle Status
- - Current Stage: Draft
- Next Review Date: 2026-03-06
- Known Issues: None
- Planned Improvements:
- Performance optimization
- Additional feature support
技能:资金趋势预测器
ID: 200
版本: 1.0.0
作者: OpenClaw Agent
许可证: MIT
概述
资金趋势预测器是一款智能分析工具,利用自然语言处理(NLP)技术分析全球主要科研资助机构(NIH、NSF、Horizon Europe)的已资助项目摘要,并预测未来3-5年的资助偏好变化趋势。
功能特点
- - 多源数据采集:自动从NIH、NSF、Horizon Europe获取已资助项目数据
- NLP深度分析:使用先进的文本挖掘技术提取主题、关键词和研究趋势
- 趋势预测模型:基于时间序列分析和主题建模预测资助方向变化
- 可视化报告:生成图表和趋势报告,直观展示分析结果
- 领域细分:按医学、工程、自然科学等领域进行分类分析
安装
bash
进入技能目录
cd skills/funding-trend-forecaster
安装依赖
pip install -r requirements.txt
下载NLTK数据
python -c import nltk; nltk.download(punkt); nltk.download(stopwords); nltk.download(wordnet)
依赖项
requests>=2.28.0
beautifulsoup4>=4.11.0
pandas>=1.5.0
numpy>=1.23.0
scikit-learn>=1.1.0
textblob>=0.17.1
nltk>=3.7
matplotlib>=3.6.0
seaborn>=0.12.0
wordcloud>=1.8.0
python-dateutil>=2.8.0
使用方法
命令行界面
bash
运行完整分析工作流
python scripts/main.py --analyze-all --output report.json
仅分析特定机构
python scripts/main.py --source nih --months 6
生成可视化报告
python scripts/main.py --visualize --input data.json --output charts/
查看趋势预测
python scripts/main.py --forecast --years 5 --output forecast.json
API调用
python
from scripts.main import FundingTrendForecaster
初始化预测器
forecaster = FundingTrendForecaster()
采集数据
forecaster.collect
data(sources=[nih, nsf, horizoneurope], months=6)
执行分析
results = forecaster.analyze_trends()
生成预测
forecast = forecaster.predict_trends(years=5)
导出报告
forecaster.export
report(outputpath=report.pdf, format=pdf)
参数说明
| 参数 | 类型 | 默认值 | 必填 | 描述 |
|---|
| --analyze-all | 标志 | false | 否 | 对所有数据源运行完整分析工作流 |
| --source |
字符串 | - | 否 | 要分析的特定机构(nih, nsf, horizon_europe) |
| --months | 整数 | 6 | 否 | 要分析的历史数据月份数 |
| --years | 整数 | 5 | 否 | 趋势预测的年数 |
| --visualize | 标志 | false | 否 | 生成可视化图表 |
| --forecast | 标志 | false | 否 | 生成趋势预测 |
| --input, -i | 字符串 | - | 否 | 输入数据文件路径(用于可视化/预测) |
| --output, -o | 字符串 | - | 否 | 输出文件路径 |
| --config | 字符串 | config.json | 否 | 配置文件路径 |
数据源
| 机构 | 数据源URL | 更新频率 |
|---|
| NIH | https://reporter.nih.gov/ | 每日 |
| NSF |
https://www.nsf.gov/awardsearch/ | 每日 |
| Horizon Europe | https://ec.europa.eu/info/funding-tenders/opportunities/ | 每周 |
配置
创建 config.json 文件以自定义分析参数:
json
{
sources: {
nih: {
enabled: true,
base_url: https://reporter.nih.gov/,
max_results: 1000
},
nsf: {
enabled: true,
base_url: https://www.nsf.gov/awardsearch/,
max_results: 1000
},
horizon_europe: {
enabled: true,
base_url: https://ec.europa.eu/info/funding-tenders/,
max_results: 500
}
},
nlp: {
language: en,
minwordlength: 3,
max_topics: 20,
stop_words: [research, study, project]
},
forecast: {
method: lda_trend,
confidence_level: 0.95,
years_ahead: 5
}
}
输出格式
JSON报告结构
json
{
metadata: {
generated_at: 2024-01-15T10:30:00Z,
data_period: 2023-07-01 to 2024-01-01,
sources: [nih, nsf, horizon_europe],
total_projects: 15420
},
trend_analysis: {
top_keywords: [
{term: artificial intelligence, frequency: 342, growth: 0.45},
{term: climate change, frequency: 298, growth: 0.32}
],
emerging_topics: [
{topic: Large Language Models, projects: 89, trend: rising},
{topic: Carbon Capture, projects: 156, trend: stable}
],
funding_shifts: {
increasing: [AI/ML, Climate Tech, Quantum Computing],
decreasing: [Traditional Materials, Fossil Fuels Research]
}
},
forecast: {
2025: {
predictedhottopics: [Generative AI, Gene Editing, Fusion Energy],
confidence: 0.87
},
2026-2029: {
longtermtrends: [AGI Safety, Personalized Medicine, Space Mining],
confidence: 0.72
}
}
}
架构
funding-trend-forecaster/
├── scripts/
│ ├── main.py # 主入口
│ ├── collectors/ # 数据采集模块
│ │ ├── init.py
│ │ ├── nih_collector.py
│ │ ├── nsf_collector.py
│ │ └── horizon_collector.py
│ ├── analyzers/ # NLP分析模块
│ │ ├── init.py
│ │ ├── text_processor.py
│ │ ├── topic_modeler.py
│ │ └── trend_detector.py
│ ├── predictors/ # 预测模块
│ │ ├── init.py
│ │ └── trend_forecaster.py
│ └── utils/ # 工具模块
│ ├── init.py
│ ├── config.py
│ └── visualizer.py
├── data/ # 数据存储
│ ├── raw/
│ └── processed/
├── output/ # 输出目录
├── config.json # 配置文件
├── requirements.txt # Python依赖项
└── SKILL.md # 本文档
发展路线图
- - [x] 基础架构设计
- [x] 核心分析模块
- [ ] 更多数据源支持(Wellcome Trust、JSPS等)
- [ ] 实时数据流处理
- [ ] 交互式Web界面
- [ ] 机器学习模型优化
许可证
MIT许可证 - 详见项目根目录下的LICENSE文件
由OpenClaw Agent生成 | 技能ID: 200
风险评估
| 风险指标 | 评估 | 等级 |
|---|
| 代码执行 | 带工具的Python脚本 | 高 |
| 网络访问 |
外部API调用 | 高 |
| 文件系统访问 | 读写数据 | 中 |
| 指令篡改 | 标准提示词指南 | 低 |
| 数据泄露 | 数据安全处理 | 中 |
安全检查清单