Information Extraction
- 16 installs
- 869 repo stars
- Updated June 8, 2026
- beita6969/scienceclaw
information-extraction is a Claude skill that extracts structured entities, relations, events, and clauses from unstructured scientific and legal documents.
About
information-extraction is a Claude skill for pulling structured entities, relations, and clauses out of unstructured scientific and domain-specific text. A developer uses it to run named-entity recognition, relation extraction, and ontology normalization on documents like biomedical papers or legal contracts. It matters because it follows a defined protocol and grounds itself in standard datasets and identifiers (MeSH, ChEBI, UniProt) rather than free-form extraction.
- Extracts structured entities, relations, events, and clauses from scientific and legal text
- 6-step protocol from schema definition through NER, relation extraction, and ontology normalization
- References benchmark datasets: ChemProt, CUAD, JNLPBA, SciERC
Information Extraction by the numbers
- 16 all-time installs (skills.sh)
- Ranked #1,318 of 2,065 Data Science & ML skills by installs in the Skillselion catalog
- Data as of Aug 2, 2026 (Skillselion catalog sync)
information-extraction capabilities & compatibility
Free; a prompt/protocol skill with no external dependencies or keys.
- Capabilities
- entity extraction · relation extraction · clause extraction
- Use cases
- data analysis · research
- Pricing
- Free
What information-extraction says it does
Extract structured entities, relations, and clauses from scientific and legal documents
CUAD** (atticus-project/cuad): Contract Understanding Atticus Dataset; 41 clause types from 510 legal contracts (CC-BY licensed)
Provide span offsets for traceability back to source text
npx skills add https://github.com/beita6969/scienceclaw --skill information-extractionAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 16 |
|---|---|
| repo stars | ★ 869 |
| Last updated | June 8, 2026 |
| Repository | beita6969/scienceclaw ↗ |
What it does
Extract structured entities, relations, and clauses from scientific and legal documents with ontology normalization.
Who is it for?
Structured extraction of chemical-protein interactions, gene-disease associations, and legal clauses from documents.
Skip if: General summarization or extraction without a defined schema; the protocol depends on target entity and relation types.
When should I use this skill?
The user needs to pull structured entities, relations, or contract clauses out of scientific or legal text.
What you get
Structured JSON, RDF triples, or a knowledge graph of normalized entities and relations with span offsets.
- Structured JSON, RDF triples, or a knowledge graph
- Extracted entities with span offsets and confidence scores
By the numbers
- 6-step extraction protocol
- CUAD covers 41 clause types from 510 legal contracts
- ChemProt has 10 relation types (CPR:3-CPR:9)
Files
Information Extraction
Purpose
Extract structured information (entities, relations, events, clauses) from unstructured scientific and domain-specific text.
Key Datasets
- ChemProt (bigbio/chemprot): Chemical-protein interaction extraction from BioCreative VI; 10 relation types (CPR:3-CPR:9) between chemicals and proteins
- CUAD (atticus-project/cuad): Contract Understanding Atticus Dataset; 41 clause types from 510 legal contracts (CC-BY licensed)
- JNLPBA: Biomedical named entity recognition (protein, DNA, RNA, cell line, cell type)
- SciERC: Scientific entity and relation extraction from AI paper abstracts
Protocol
1. Schema definition — Define target entity types, relation types, and attributes 2. Preprocessing — Sentence segmentation, tokenization, abbreviation expansion 3. Entity recognition — Identify and classify named entities (NER) 4. Relation extraction — Detect relationships between entity pairs (RE) 5. Normalization — Map entities to standard ontologies (MeSH, ChEBI, UniProt) 6. Output structuring — Format as structured JSON, RDF triples, or knowledge graph
Extraction Types
- Chemical-protein interactions: Substrate, inhibitor, agonist, antagonist, activator
- Legal clause extraction: Termination, IP rights, non-compete, indemnification, limitation of liability
- Gene-disease associations: Causal, biomarker, therapeutic target
- Drug-drug interactions: Synergistic, antagonistic, pharmacokinetic
Rules
- Report extraction confidence scores for each entity/relation
- Provide span offsets for traceability back to source text
- Normalize entities to standard identifiers (CAS, UniProt ID, etc.)
- Handle nested entities and overlapping relations
- Validate extracted facts against known databases when possible
Related skills
Microsoft FoundryDeploy, evaluate, and continuously improve Microsoft Foundry agents from a single agent interface.490k1.3k
Ai Research ReproductionOrchestrate trustworthy, auditable reproduction of deep learning repositories directly from their READMEs.176k511Run TrainSafely execute selected deep learning training commands with standardized evidence capture.176k511Explore RunSafely run isolated exploratory experiments with clear recording and conservative selection before committing changes.176k511Paper Context ResolverFetch precise reproduction-critical details like dataset splits, preprocessing steps, or evaluation protocols from the original academic paper when the repo README leav141k511Repo Intake And PlanScan unfamiliar AI research repositories and receive a minimal, trustworthy reproduction target before investing significant time.140k511FAQ
Which datasets does this skill reference?
ChemProt for chemical-protein interactions, CUAD for legal contract clauses, JNLPBA for biomedical NER, and SciERC for scientific entity and relation extraction.
How is traceability handled?
The rules require providing span offsets for traceability back to source text and normalizing entities to standard identifiers like CAS and UniProt IDs.