extract
Extract content from specific URLs using Tavily's extraction API. Returns clean markdown/text from web pages. Use when you have specific URLs and need their content without writing code.
81
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillEvaluation — 88%
↑ 2.66xAgent success when using this skill
Validation for skill structure
Evaluation results
100%
70%Research Content Aggregator
Batching and targeted extraction
Uses extract.sh script
0%
100%
JSON argument format
0%
100%
Batches into multiple requests
0%
100%
Uses query parameter
0%
100%
Uses chunks_per_source
0%
100%
chunks_per_source paired with query
100%
100%
Reports failed results
100%
100%
Markdown format preserved
100%
100%
Without context: $0.1838 · 52s · 12 turns · 16 in / 3,348 out tokens
With context: $0.4291 · 1m 25s · 19 turns · 2,438 in / 5,042 out tokens
65%
-5%Competitor Pricing Monitor
JS-page depth and timeout handling
Uses extract.sh script
0%
0%
Selects advanced depth
100%
50%
Sets extended timeout
0%
100%
Timeout within valid range
100%
100%
All 6 URLs included
100%
100%
Checks failed_results
100%
100%
JSON argument format
100%
0%
Without context: $0.2077 · 1m 2s · 15 turns · 21 in / 3,403 out tokens
With context: $0.4716 · 1m 43s · 23 turns · 265 in / 5,649 out tokens
100%
100%Product Catalog Content Harvester
Image extraction and response handling
Uses extract.sh script
0%
100%
Enables image extraction
0%
100%
Uses query parameter
0%
100%
Uses chunks_per_source
0%
100%
chunks_per_source value is 3
0%
100%
Plain text format
0%
100%
Extracts failed_results
0%
100%
JSON argument format
0%
100%
Without context: $0.8143 · 4m 33s · 30 turns · 34 in / 14,479 out tokens
With context: $0.4404 · 1m 32s · 24 turns · 25 in / 4,935 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Unknown
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.