---
name: chematagent-drug-discovery
source: https://app.decimal.ai/s/chematagent-drug-discovery@2/SKILL.md
source_sha256: 80c456e3086d
---

<!--
# COPYRIGHT NOTICE
# This file is part of the "Universal Biomedical Skills" project.
# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
# All Rights Reserved.
#
# This code is proprietary and confidential.
# Unauthorized copying of this file, via any medium is strictly prohibited.
#
# Provenance: Authenticated by MD BABU MIA

-->

# CheMatAgent

A two-tiered agent system with access to 137 Python-wrapped chemical tools for drug discovery and materials science.

## When to Use

*   **Molecule Design**: Generating novel structures with specific properties.
*   **Property Prediction**: Estimating solubility, toxicity, and bioactivity.
*   **Synthesis Planning**: Designing retro-synthetic routes.

## Core Capabilities

1.  **Tool Orchestration**: Manages a library of 137 chemical tools.
2.  **Multi-Scale Modeling**: Bridges quantum mechanics and molecular dynamics.
3.  **Lab Automation**: Generates instructions for robotic synthesis platforms.

## Workflow

1.  **Goal**: Define target property (e.g., "LogP < 5").
2.  **Design**: Generate candidates.
3.  **Filter**: Use property prediction tools.
4.  **Plan**: Output synthesis recipe.

## Example Usage

**User**: "Design a molecule similar to Aspirin but with higher solubility."

**Agent Action**:
```bash
python -m chematagent.design --scaffold "Aspirin" --objective "maximize solubility"
```


<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->