sk Skill · affaan-m
regex-vs-llm-structured-text
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
Open on skills.sh ↗read 2026-09-17
- installs 8w
- 1,514
- 30-day movement
- starts with the next reading
- Related entries
- 6
- Connections
- 0
pythonJavaScript
An agent skill that provides a decision framework for choosing between regex and LLM approaches when parsing structured text. It prescribes a hybrid pipeline: regex handles the majority of items, confidence scoring flags uncertain cases, and an LLM validates only those edge cases.
Reach for it when you need to parse structured text without over-spending on LLM calls or under-serving variable inputs.
Use it to
- Parse quiz or exam questions
- Extract form data
- Process invoices and receipts
- Parse document structure like headers and tables
- Build a hybrid regex-plus-LLM extraction pipeline
For Developers building text extraction pipelines with AI coding agents
- Host repository
- affaan-m/ECC
- Installs, lifetime
- 2,800
- Installs, 8 weeks
- 1,514
- Host stars
- 261k
- Host language
- JavaScript
topicsregexllmtext-parsingstructured-texthybrid-pipelinedecision-framework