Skills for Real Engineers. Straight from my .agents directory.
39 skills
esmfold2
Agent skill from
Biohub ESMFold2 / ESMFold2-Fast all-atom co-folding (Candido et al. 2026, github.com/Biohub/esm). Single-sequence and MSA modes; protein, DNA, RNA, ligand (CCD/SMILES), modified residues. FoldBench Ab-Ag 50-55%, PPI 70-77% DockQ-pass. Also covers the ESMC-{300M,600M,6B} protein language models from the same release: masked-LM logits, hidden states, mutation scoring, contact prediction, and the SAE interpretability head. MIT-licensed weights on HuggingFace org `biohub`. Use this skill when: (1) Predicting complex structures with single-sequence input, (2) Validating designed binders with ESMFold2-Fast, (3) Running ESMFold2 with MSA input, (4) Getting ESMC embeddings or per-residue mutation scores, (5) Choosing kernel backend and sampling-step settings for paper-faithful throughput.
How to install esmfold2
install esmfold2
# install the open-science pack with the skills CLI
npx skills add aipoch/open-science# or add only this skill to Claude Code: copy the folder
resources/skills/esmfold2/
# into
~/.claude/skills/esmfold2/ (all projects)
.claude/skills/esmfold2/ (one project)
1# ESMFold2 (Biohub)23All-atom diffusion co-folding from the Biohub ESM release (2026). ESMFold2 =448 pair layers with MSA support; ESMFold2-Fast = 24 layers, single-sequence5only, ~1.7x faster.67**License:** MIT (code github.com/Biohub/esm + weights HF `biohub/*`).8**Paper:** "Language Modeling Materializes a World Model of Protein Biology" (2026).910## Install1112CUDA 12.x GPU (H100/A100-class); Python **3.12 only**. Fresh venv; needs13egress to HF Hub, GitHub, PyPI:1415```bash16pip install --no-cache-dir uv17uv venv --python 3.12 /work/venv && source /work/venv/bin/activate18uv pip install \19 "torch>=2.5,<2.8" einops "biotite>=1.0" rdkit msgpack-numpy biopython \20 scikit-learn brotli attrs pandas cloudpathlib httpx tenacity zstd pydssp \21 pygtrie accelerate huggingface_hub safetensors "numpy<3" networkx \22 sentencepiece tokenizers regex packaging filelock pyyaml typing_extensions \23 "transformers @ git+https://github.com/Biohub/transformers.git@3a8956fb4d4ea16b0ec8e71deef2c2909b6a5cbf"24uv pip install --no-deps "esm @ git+https://github.com/Biohub/esm.git@f652b471"25# OPTIONAL — only affects ESMC attention; trunk speedup comes from set_kernel_backend("fused")26uv pip install ninja packaging wheel setuptools27MAX_JOBS=8 uv pip install --no-deps --no-build-isolation "flash-attn<3"28# Do NOT install transformer-engine — RuntimeError (not ImportError) on import29# slips ESMC's guard and kills ESMFold2Model import.30```31…
Excerpt shown under the repository's Apache-2.0 license.
More skills in open-science
Keep exploring
Similar skill repositories
An agentic skills framework & software development methodology that works.
15 skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
CC1 skill
Public repository for Agent Skills
20 skills
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
CCCU6 skills
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
CCCXGHGM+135 skills