I built a scraping engine where the LLM writes configs, not code

developer-tools / Show HN
45
SCORE

Fitter is a scraping engine that uses LLM-generated configuration files rather than custom code to define data extraction logic. Users describe what they want to scrape and the LLM produces the config, which Fitter then executes against target pages. The project is open source and available on GitHub.

Sources (1)

Score Breakdown

Traction
raw 1.00 · weight 35%
3.5pts
Novelty
0 days old · weight 20%
20.0pts
Source diversity
1 source · weight 10%
3.3pts
AI quality
raw 52.00 · weight 35%
18.2pts

Interesting LLM-config approach to scraping, but zero traction signals and unclear differentiation from existing scraping frameworks limit confidence.

Final Score45/100

Similar Products

omnigent
AI-TOOLS · GITHUB
84
SCORE
devspace
DEVELOPER-TOOLS · GITHUB
84
SCORE
eve
DEVELOPER-TOOLS · GITHUB
84
SCORE
coreai-models
AI-TOOLS · GITHUB
84
SCORE
kage
DEVELOPER-TOOLS · GITHUB
84
SCORE
Marzban-Node
INFRASTRUCTURE · GITHUB
84
SCORE
SUPPORT THE PROJECT →
← ALL PRODUCTS© 2026 MARKETHUNT