funding-trend-forecaster
Predict funding trend shifts using NLP analysis of grant abstracts from NIH, NSF, and Horizon Europe
54
34%
Does it follow best practices?
Impact
84%
4.42xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./scientific-skills/Academic Writing/funding-trend-forecaster/SKILL.mdEvaluation results
87%
67%Multi-Agency Funding Trend Analysis and Forecast
CLI parameters and JSON output structure
CLI entry point
0%
80%
All-sources flag
0%
100%
Months parameter
0%
100%
Forecast flag
0%
100%
Forecast years parameter
0%
100%
Output flag
0%
80%
Report metadata section
0%
0%
Metadata sources list
0%
100%
Metadata data period
0%
100%
Top keywords section
100%
100%
Forecast section present
100%
100%
Forecast confidence values
0%
100%
Without context: $1.3508 · 9m 52s · 47 turns · 53 in / 17,658 out tokens
With context: $0.6149 · 1m 42s · 27 turns · 31 in / 4,500 out tokens
100%
72%Targeted Funding Analysis with Custom NLP Configuration
Custom configuration file structure
Top-level sources key
0%
100%
NIH source enabled
0%
100%
Horizon Europe source enabled
0%
100%
NSF source disabled
0%
100%
Source base_url fields
0%
100%
Source max_results fields
0%
100%
NLP section present
100%
100%
NLP max_topics set to 30
0%
100%
NLP stop_words list
100%
100%
Forecast section present
100%
100%
Forecast method field
0%
100%
Config flag used
100%
100%
NIH base_url correct
0%
100%
Without context: $2.8365 · 11m 34s · 87 turns · 2,344 in / 29,393 out tokens
With context: $0.7622 · 2m 20s · 32 turns · 35 in / 6,587 out tokens
65%
55%Automated Funding Intelligence Pipeline
Python API programmatic usage
Imports from scripts.main
0%
100%
FundingTrendForecaster instantiated
0%
100%
collect_data called
0%
0%
Correct source names in collect_data
0%
0%
Months parameter in collect_data
0%
0%
analyze_trends called
0%
100%
predict_trends called with years
0%
100%
export_report called
0%
100%
Correct method call order
0%
50%
funding_report.json produced
100%
100%
Without context: $1.2361 · 5m 1s · 54 turns · 989 in / 17,175 out tokens
With context: $0.7899 · 2m 37s · 31 turns · 204 in / 8,155 out tokens
- Repository
- aipoch/medical-research-skills
- Commit
4a48721
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.