Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/chemeagle-guide
Multi-agent system for chemical literature information extraction
Passed 23 of 23 test cases with the skill, and 10 of 23 without it.
and 19 more in the full report
13 fixed10 passed either way0 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 13 test cases the skill fixed. Same model, same prompt, with and without the skill.
When searching a ChemEagle ReactionDatabase, each hit object contains literature metadata. What property name on a hit object stores the source paper DOI?
| Checked | Without skill | With skill |
|---|---|---|
| The response identifies paper_doi as the attribute name for search hit literature DOI. | Missed | Passed |
The instructions your model reads when the skill loads.
ChemEagle is a multi-agent system for extracting structured chemical information from scientific literature. It uses specialized agents for recognizing chemical entities, extracting reaction conditions, identifying product yields, and building structured databases from unstructured chemistry papers. Particularly useful for building reaction databases and automating systematic reviews in chemistry.
Chemistry Paper (PDF/text)
↓
Document Parser Agent (section identification)
↓
Chemical NER Agent
├── Compound names → SMILES/InChI
├── Reagents and catalysts
├── Solvents and conditions
└── Product identification
↓
Reaction Extraction Agent
├── Reactants → Products mapping
├── Reaction conditions (T, P, time)
├── Yields and selectivity
└── Procedure steps
↓
Validation Agent (cross-check extracted data)
↓
Structured Output (JSON, CSV, database)
from chemeagle import ChemEagle
eagle = ChemEagle(llm_provider="anthropic")
# Extract from a chemistry paper
result = eagle.extract("paper.pdf")
test cases passed, vs 10 / 23 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.