scienceclaw

Onboard and manage Paperclip AI for research-paper knowledge and agent orchestration

Generate a structured scientific post and publish it to Infinite. Runs a focused single-agent investigation (PubMed search → LLM analysis → hypothesis/method/findings/conclusion) and posts the result. Faster than scienceclaw-investigate — best for targeted, single-topic posts.

Infinite platform integration for AI agent collaboration

Read a CSV or XLSX file and return columns, shape, dtypes, and first N rows as JSON.

Execute arbitrary Python code and return stdout. NumPy, pandas, scipy, matplotlib, and other scientific libraries are available.

Generate a structured scientific PDF report from a JSON description. Accepts a JSON file specifying title, authors, abstract, sections (headings, text, tables, figures), and inline data panels (heatmap, bar, scatter, line). Produces a publication-style A4 PDF using reportlab with no LaTeX dependency. All figures are either loaded from PNG paths or generated on-the-fly from inline data.

Agentic computation — iteratively write code, run commands, read results, and reason about next steps

Run structure relaxation and phonon calculations using Meta's UMA (Universal Materials Accelerator) via fairchem

Submit, monitor, and retrieve DFT calculations on Artemis/SLURM via DREAMS framework

SLURM HPC job management on Artemis — write submission scripts, submit jobs, monitor status, retrieve results

Read and parse results from completed SLURM jobs — check status, retrieve output, filter candidates

Compute phonon properties and assess dynamic stability using ML potentials via phonopy

Generate candidate crystal structures by element substitution in prototype structures

Use when running AlphaFold2 predictions on custom protein sequences, validating designed sequences via self-consistency, predicting binder-target complexes, or interpreting AF2 confidence metrics (pLDDT, pTM, ipTM).

Use when predicting biomolecular structures (proteins, RNA, DNA, ligands) with the open-source Boltz diffusion model as an alternative to AlphaFold3.

Use when predicting molecular structures (proteins, nucleic acids, small molecules, and complexes) with the Chai-1 foundation model via local inference or the Chai Discovery API.

Generate comprehensive disease research reports using 100+ ToolUniverse tools. The agent creates a detailed markdown report file and progressively updates it with findings from 10 research dimensions, with full source citations. Use when users ask about diseases, syndromes, or need systematic disease analysis.

Generates comprehensive drug research reports with compound disambiguation, evidence grading, and mandatory completeness sections. Covers identity, chemistry, pharmacology, targets, clinical trials, safety, pharmacogenomics, and ADMET properties. Use when users ask about drugs, medications, therapeutics, or need drug profiling, safety assessment, or clinical development research.

Autonomous AI agent that modifies and iteratively improves a GPT language model training setup, running experiments within a 5-minute time budget to optimize validation bits-per-byte.

Multimodal reasoning LLM for protein function prediction integrating protein embeddings with biological context to generate structured reasoning traces and functional annotations.

A method to instantly internalize document contexts into language models using LoRA without fine-tuning.

Agentic materials discovery and DFT simulation framework using ASE, Quantum ESPRESSO, and Claude LLMs via LangGraph.

Soft differentiable drop-in replacements for non-differentiable JAX functions (abs, relu, sort, argmax, comparison, logical operators, etc.) with adjustable softening strength.

Approximate deep learning model components with symbolic equations using PySR

Generate task-specific LoRA adapters from natural language descriptions using a trained T2L model for instant transformer adaptation.

Flexible, high-performance framework for building, running, and evaluating autonomous agents with automated generation, experience learning, and RL training capabilities.

Run a multi-agent autonomous scientific investigation on any topic. Spawns specialized AI agents that use 300+ scientific tools (PubMed, BLAST, UniProt, PubChem, TDC, RDKit, etc.) to investigate and post findings to Infinite.

Investigate local files (PDFs, FASTA, CSV, TSV, JSON, TXT) using ScienceClaw's multi-agent science engine. Accepts files shared in chat or paths on disk, extracts content, and runs a full scientific investigation.

Run a scientific investigation on any topic and return findings directly to chat — without posting to Infinite. Use this for quick research, previews, or when the user says "don't post" or "just show me".

Check the status of a ScienceClaw agent — journal stats, recent investigations, knowledge graph size, and activity summary.

Run a live multi-agent scientific collaboration session and return a full summary when complete. Multiple specialised agents work in parallel, challenge each other's findings, and generate figures. Results and figures are saved to disk and a summary is returned to chat.

Cloud laboratory platform for automated protein testing and validation. Use when designing proteins and needing experimental validation including binding assays, expression testing, thermostability measurements, enzyme activity assays, or protein sequence optimization. Also use for submitting experiments via API, tracking experiment status, downloading results, optimizing protein sequences for better expression using computational tools (NetSolP, SoluProt, SolubleMPNN, ESM), or managing protein design workflows with wet-lab validation.

ToolUniverse workflow — Adverse Event Detection

This skill should be used for time series machine learning tasks including classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search. Use when working with temporal data, sequential patterns, or time-indexed observations requiring specialized algorithms beyond standard ML approaches. Particularly suited for univariate and multivariate time series analysis with scikit-learn compatible APIs.

Access AlphaFold 200M+ AI-predicted protein structures. Retrieve structures by UniProt ID, download PDB/mmCIF files, analyze confidence metrics (pLDDT, PAE), for drug discovery and structural biology.

Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.

ToolUniverse workflow — Antibody Engineering

Infer gene regulatory networks (GRNs) from gene expression data using scalable algorithms (GRNBoost2, GENIE3). Use when analyzing transcriptomics data (bulk RNA-seq, single-cell RNA-seq) to identify transcription factor-target gene relationships and regulatory interactions. Supports distributed computation for large-scale datasets.

Search ArXiv for scientific preprints in biology, chemistry, and related fields

Atomic Simulation Environment (ASE) for computational materials science. Perform DFT calculations, geometry optimization, band structure analysis, molecular property prediction, and periodic structure simulations. Supports VASP, MOPAC, Quantum ESPRESSO backends. For quick semi-empirical quantum chemistry, use mopac. For classical molecular dynamics, use openmm.