install
source · Clone the upstream repo
git clone https://github.com/FreedomIntelligence/OpenClaw-Medical-Skills
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/FreedomIntelligence/OpenClaw-Medical-Skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/antibody-design-agent/MAGE" ~/.claude/skills/freedomintelligence-openclaw-medical-skills-mage && rm -rf "$T"
OpenClaw · Install into ~/.openclaw/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/FreedomIntelligence/OpenClaw-Medical-Skills "$T" && mkdir -p ~/.openclaw/skills && cp -r "$T/skills/antibody-design-agent/MAGE" ~/.openclaw/skills/freedomintelligence-openclaw-medical-skills-mage && rm -rf "$T"
manifest:
skills/antibody-design-agent/MAGE/SKILL.mdsource content
<!--
# 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
-->
name: mage-antibody-generator description: Ab seq forge keywords:
- antibody
- antigen
- FASTA
- generation
- validation measurable_outcome: Generate the requested number of antibody sequences (default ≥5) with metadata (model checkpoint, seed) and deliver FASTA files within 10 minutes. license: MIT metadata: author: MAGE Team version: "1.0.0" compatibility:
- system: Python 3.9+ / GPU allowed-tools:
- run_shell_command
- read_file
MAGE (Monoclonal Antibody Generator)
Run the MAGE antibody generation workflow to propose antigen-conditioned antibody sequences for downstream structural validation.
Workflow
- Prep env:
and install dependencies, then point to GPU if available.cd repo - Run generator:
.python generate_antibodies.py --antigen_sequence <SEQ> --num_candidates N --output_dir ./results - Collect outputs: Provide FASTA paths + metadata, optionally translate into JSON manifest.
- Recommend validation: Suggest AlphaFold/Rosetta checks and wet-lab follow-up.
Guardrails
- Never imply binding efficacy without structural/experimental confirmation.
- Track model version + seeds to ensure reproducibility.
- Encourage downstream filtering (liability motifs, developability metrics).
References
- Source instructions in
and repo scripts.README.md