prompt
FreeScientific database orchestrator and molecular research agent system prompt
FreeFree tier
Inputs: textOutputs: text
About prompt
This is a system prompt designed for an AI agent that acts as a scientific database orchestrator and molecular research agent. It provides structured guidelines for querying, integrating, and verifying data across major repositories in structural biology, cheminformatics, genomics, proteomics, and scholarly literature. The prompt enforces operational principles such as wrapper-first execution (using provided helper scripts instead of direct REST calls), identifier resolution before querying, rate-limit compliance, license notification, and fact verification over parametric knowledge. It covers databases like AlphaFold DB, RCSB PDB, UniProt, ChEMBL, OpenTargets, ClinicalTrials.gov, PubMed, Ensembl, Reactome, and many others.
Key Features
Wrapper-first execution using helper scripts for rate limits and audit
Identifier resolution before querying (e.g., UniProt, CID, ENSEMBL, DOI)
Covers AlphaFold DB, RCSB PDB, UniProt, ChEMBL, PubChem, OpenTargets, ClinVar, ClinicalTrials.gov, PubMed, Europe PMC, OpenAlex, bioRxiv, arXiv, Ensembl, Reactome, KEGG and more
Rate-limit and terms-of-service compliance with automatic pause and escalation
License notification on first use of any database skill
Fact verification over parametric knowledge for specific scientific data points
Supports protein structure prediction, molecular docking, drug mechanism, variant interpretation, pathway enrichment, tissue expression QTLs, and literature search
Pros & Cons
Pros
- Provides structured and disciplined approach to querying multiple scientific databases
- Enforces best practices like identifier resolution and wrapper usage to reduce errors and rate-limit issues
- Licenses are checked and notified, promoting compliance
- Covers a very wide range of biomedical and molecular biology databases
- Emphasizes fact verification over relying on parametric knowledge for accuracy
Cons
- Requires user to have helper scripts/wrappers installed and configured for each database
- Not a standalone tool – it is a text prompt that must be fed into an AI chatbot that can execute code
- May need manual intervention when database wrappers return 429 or 401 errors
- User must resolve names to canonical IDs before querying, adding extra steps
Best For
Predicting protein structures using AlphaFold DB given UniProt IDsRetrieving experimental macromolecular structures and ligand binding sites from RCSB PDBAnalyzing chemical compounds, bioactivities, and ADMET properties via ChEMBL/PubChemExploring target-disease associations and pathogenic variant interpretations from OpenTargets and ClinVarSearching clinical trial statuses and regulatory labels on ClinicalTrials.gov and OpenFDAConducting literature searches with citation metrics and open-access PDF retrieval from PubMed, Europe PMC, bioRxiv, arXivPerforming pathway enrichment and controlled-vocabulary lookups using Reactome, KEGG, and Gene Ontology
FAQ
What databases does this prompt cover?
It covers AlphaFold DB, RCSB PDB, UniProt/InterPro/Pfam, ChEMBL/PubChem, OpenTargets/ClinVar/gnomAD/GTEx, ClinicalTrials.gov/OpenFDA, PubMed/Europe PMC/OpenAlex/bioRxiv/arXiv, AlphaGenome/Ensembl/dbSNP, and Reactome/KEGG/Gene Ontology (QuickGO/EBI OLS).
What are the operational principles enforced by this prompt?
1) Wrapper-first execution (use helper scripts, not raw HTTP). 2) Identifier resolution before querying. 3) Rate-limit and TOS compliance. 4) License notification on first use per session. 5) Fact verification over parametric knowledge.
Is this a standalone tool or just a prompt?
This is a system prompt (text instructions) designed for an AI agent such as a large language model. It is not a standalone application; it requires an environment that can execute code and call the mentioned database wrappers.
What kind of queries is this prompt designed for?
It is designed for scientific queries involving protein structures, chemical compounds, gene variants, clinical trials, literature, and other molecular biology and bioinformatics data.