[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M74MA3ETSVH3DCCAA7MRNTWD":3,"skill-same:01M74MA3ETSVH3DCCAA7MRNTWD":59},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":36,"commitSha":37,"skillDirectory":38,"forksCount":39,"repositoryHtmlUrl":40,"repositoryHomepageUrl":41,"ownerHtmlUrl":42,"ownerBlogUrl":6,"ownerDescription":43,"updatedAt":44,"install":45,"resources":58},"01M74MA3ETSVH3DCCAA7MRNTWD","fair-esm2",null,"Embed proteins with Meta AI's ESM-2 (`fair-esm` package). Use this skill when: (1) Extracting per-residue or per-sequence embeddings for downstream ML, (2) Masked-LM likelihood \u002F mutation effect scoring, (3) Contact prediction from a sequence.","pengqianhan\u002FAI-Human-Research-OS","pengqianhan","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F4286531?v=4","01M74M3DT9HNZ6HPAK8TX69EX9","PengqianHan","01M74M3DTVSN2GXW0JH2W396FK",3,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fclaude-science-skills\u002Fskills\u002Ffair-esm2\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: fair-esm2\ndescription: >\n  Embed proteins with Meta AI's ESM-2 (`fair-esm` package). Use this skill\n  when: (1) Extracting per-residue or per-sequence embeddings for downstream\n  ML, (2) Masked-LM likelihood \u002F mutation effect scoring, (3) Contact\n  prediction from a sequence.\nlicense: Apache-2.0\ncategory: biomodels\nrequirements: [gpu]\nmetadata:\n  display-name: ESM-2\n  # github.com\u002Ffacebookresearch\u002Fesm\u002Fblob\u002Fmain\u002FLICENSE: MIT (© Meta Platforms,\n  # Inc. and affiliates). verified 2026-06-30\n  third_party:\n    - kind: weights\n      name: ESM-2\n      provider: Meta AI\n      license: MIT\n      terms_url: https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fesm\u002Fblob\u002Fmain\u002FLICENSE\n---\n\n# fair-esm2 — ESM-2 (Meta AI)\n\nESM-2 code and weights are MIT (Meta AI, github.com\u002Ffacebookresearch\u002Fesm).\n\n> **Package disambiguation.** `pip install fair-esm` gives you `import esm`\n> with `esm.pretrained.*` (ESM-1\u002F2). Biohub's github.com\u002FBiohub\u002Fesm fork\n> (MIT) gives you `from esm.models.esmfold2 import ESMFold2InputBuilder` —\n> see the **`esmfold2`** skill. Both share the `esm` namespace but are\n> different libraries. This skill covers **fair-esm** (the Meta package).\n\n## Prerequisites\n\n| Requirement | Minimum | Recommended |\n| ----------- | ------- | ----------- |\n| Python      | 3.8+    | 3.11        |\n| CUDA        | 11.7+   | 12.x        |\n| GPU VRAM    | 8 GB (8M), 16 GB (650M) | 24 GB+ (650M \u002F 3B) |\n\n## How to run\n\n### Embeddings\n\n```python\nimport torch, esm\n\nmodel, alphabet = esm.pretrained.esm2_t33_650M_UR50D()\nmodel = model.eval().cuda()\nbc = alphabet.get_batch_converter()\n\n_, _, toks = bc([(\"ubq\", \"MQIFVKTLTGKTITLEVEPSDTIENVK\")])\nwith torch.no_grad():\n    out = model(toks.cuda(), repr_layers=[33])\nemb = out[\"representations\"][33]      # (1, L+2, 1280) — includes BOS\u002FEOS\nseq_emb = emb[0, 1:-1].mean(0)        # per-sequence mean\n```\n\n### Masked-LM scoring\n\n```python\nwith torch.no_grad():\n    out = model(toks.cuda(), repr_layers=[33])\nlogits = out[\"logits\"][0, 1:-1]       # (L, |vocab|)\n# WT marginal log-likelihood; for mutation scoring, mask the position and\n# compare logit[mut] − logit[wt].\n```\n\n### Contact prediction\n\n```python\nwith torch.no_grad():\n    out = model(toks.cuda(), repr_layers=[33], return_contacts=True)\ncontacts = out[\"contacts\"][0]         # (L, L)\n```\n\n## Models\n\n| Name                       | Layers | Dim  | Params | Use                        |\n| -------------------------- | ------ | ---- | ------ | -------------------------- |\n| `esm2_t6_8M_UR50D`         | 6      | 320  | 8 M    | Fast smoke \u002F tiny embeddings |\n| `esm2_t33_650M_UR50D`      | 33     | 1280 | 650 M  | Default embedding model    |\n| `esm2_t36_3B_UR50D`        | 36     | 2560 | 3 B    | Best embeddings, 24 GB+    |\n\n## Output format\n\n`out[\"representations\"][layer]` is `(B, L+2, D)`; slice `[ :, 1:-1, : ]` to\ndrop BOS\u002FEOS. `out[\"contacts\"]` (when `return_contacts=True`) is `(B, L, L)`.\n\n\n## Remote compute\n\nNeeds ≥16 GB VRAM (650M model) and either pre-cached `.pt` checkpoints or\negress to `dl.fbaipublicfiles.com`. Read\n`compute_details({provider, mode:'read'})` for an environment with `fair-esm`\nand a torch-hub weight cache, then:\n\n```python\nc = host.compute.create(provider)\njob = c.submit_job(\n    intent=\"ESM-2 650M embeddings for 200 sequences — 1×GPU, ~2 min\",\n    inputs=[\n        {\"src\": \"seqs.fasta\", \"dst_filename\": \"seqs.fasta\"},\n        {\"src\": \"embed_esm2.py\", \"dst_filename\": \"embed_esm2.py\"},\n    ],\n    command=\"python3 embed_esm2.py\",\n    environment=...,   # env name from compute_details\n    outputs=[\"embeddings.pt\"],\n    timeout_seconds=1800,\n)\nprint(job.job_id)   # cell ends here — kernel never blocks on compute\n```\n\nThen call the `wait_for_notification` brain-tool. When the\n`compute_done` notification arrives, act on its payload:\n\n```python\nsave_artifacts(payload[\"featured_files\"])   # paths under hpc\u002F\u003Cjob_id>\u002F\n```\n\nFor the full result dict (`output_files`, `remote_workdir`, …), re-enter the\nkernel: `c.attach_job(job_id).result()` then `c.close()`. See the\n`remote-compute-ssh` \u002F `remote-compute-modal` skill for the orchestration\ndetails.\n\nInside `embed_esm2.py`, set `TORCH_HOME` to the provider's torch-hub cache\nmount (path is in `compute_details`) so `esm.pretrained.*` resolves locally.\n\n\n## Troubleshooting\n\n| Symptom                                       | Cause                              | Fix                                   |\n| --------------------------------------------- | ---------------------------------- | ------------------------------------- |\n| `ModuleNotFoundError: No module named 'esm.models'` | You want Biohub's `esm` fork, not `fair-esm` | See `esmfold2` skill; this skill uses `esm.pretrained.*` |\n| Slow first call                               | Downloading weights via torch.hub  | Set `TORCH_HOME` to a cached location |\n\n---\n\n**Next**: feed embeddings to a classifier. For structure prediction, use\n`esmfold2`.\n","# fair-esm2 — ESM-2 (Meta AI)\n\nESM-2 code and weights are MIT (Meta AI, github.com\u002Ffacebookresearch\u002Fesm).\n\n> **Package disambiguation.** `pip install fair-esm` gives you `import esm`\n> with `esm.pretrained.*` (ESM-1\u002F2). Biohub's github.com\u002FBiohub\u002Fesm fork\n> (MIT) gives you `from esm.models.esmfold2 import ESMFold2InputBuilder` —\n> see the **`esmfold2`** skill. Both share the `esm` namespace but are\n> different libraries. This skill covers **fair-esm** (the Meta package).\n\n## Prerequisites\n\n| Requirement | Minimum | Recommended |\n| ----------- | ------- | ----------- |\n| Python      | 3.8+    | 3.11        |\n| CUDA        | 11.7+   | 12.x        |\n| GPU VRAM    | 8 GB (8M), 16 GB (650M) | 24 GB+ (650M \u002F 3B) |\n\n## How to run\n\n### Embeddings\n\n```python\nimport torch, esm\n\nmodel, alphabet = esm.pretrained.esm2_t33_650M_UR50D()\nmodel = model.eval().cuda()\nbc = alphabet.get_batch_converter()\n\n_, _, toks = bc([(\"ubq\", \"MQIFVKTLTGKTITLEVEPSDTIENVK\")])\nwith torch.no_grad():\n    out = model(toks.cuda(), repr_layers=[33])\nemb = out[\"representations\"][33]      # (1, L+2, 1280) — includes BOS\u002FEOS\nseq_emb = emb[0, 1:-1].mean(0)        # per-sequence mean\n```\n\n### Masked-LM scoring\n\n```python\nwith torch.no_grad():\n    out = model(toks.cuda(), repr_layers=[33])\nlogits = out[\"logits\"][0, 1:-1]       # (L, |vocab|)\n# WT marginal log-likelihood; for mutation scoring, mask the position and\n# compare logit[mut] − logit[wt].\n```\n\n### Contact prediction\n\n```python\nwith torch.no_grad():\n    out = model(toks.cuda(), repr_layers=[33], return_contacts=True)\ncontacts = out[\"contacts\"][0]         # (L, L)\n```\n\n## Models\n\n| Name                       | Layers | Dim  | Params | Use                        |\n| -------------------------- | ------ | ---- | ------ | -------------------------- |\n| `esm2_t6_8M_UR50D`         | 6      | 320  | 8 M    | Fast smoke \u002F tiny embeddings |\n| `esm2_t33_650M_UR50D`      | 33     | 1280 | 650 M  | Default embedding model    |\n| `esm2_t36_3B_UR50D`        | 36     | 2560 | 3 B    | Best embeddings, 24 GB+    |\n\n## Output format\n\n`out[\"representations\"][layer]` is `(B, L+2, D)`; slice `[ :, 1:-1, : ]` to\ndrop BOS\u002FEOS. `out[\"contacts\"]` (when `return_contacts=True`) is `(B, L, L)`.\n\n\n## Remote compute\n\nNeeds ≥16 GB VRAM (650M model) and either pre-cached `.pt` checkpoints or\negress to `dl.fbaipublicfiles.com`. Read\n`compute_details({provider, mode:'read'})` for an environment with `fair-esm`\nand a torch-hub weight cache, then:\n\n```python\nc = host.compute.create(provider)\njob = c.submit_job(\n    intent=\"ESM-2 650M embeddings for 200 sequences — 1×GPU, ~2 min\",\n    inputs=[\n        {\"src\": \"seqs.fasta\", \"dst_filename\": \"seqs.fasta\"},\n        {\"src\": \"embed_esm2.py\", \"dst_filename\": \"embed_esm2.py\"},\n    ],\n    command=\"python3 embed_esm2.py\",\n    environment=...,   # env name from compute_details\n    outputs=[\"embeddings.pt\"],\n    timeout_seconds=1800,\n)\nprint(job.job_id)   # cell ends here — kernel never blocks on compute\n```\n\nThen call the `wait_for_notification` brain-tool. When the\n`compute_done` notification arrives, act on its payload:\n\n```python\nsave_artifacts(payload[\"featured_files\"])   # paths under hpc\u002F\u003Cjob_id>\u002F\n```\n\nFor the full result dict (`output_files`, `remote_workdir`, …), re-enter the\nkernel: `c.attach_job(job_id).result()` then `c.close()`. See the\n`remote-compute-ssh` \u002F `remote-compute-modal` skill for the orchestration\ndetails.\n\nInside `embed_esm2.py`, set `TORCH_HOME` to the provider's torch-hub cache\nmount (path is in `compute_details`) so `esm.pretrained.*` resolves locally.\n\n\n## Troubleshooting\n\n| Symptom                                       | Cause                              | Fix                                   |\n| --------------------------------------------- | ---------------------------------- | ------------------------------------- |\n| `ModuleNotFoundError: No module named 'esm.models'` | You want Biohub's `esm` fork, not `fair-esm` | See `esmfold2` skill; this skill uses `esm.pretrained.*` |\n| Slow first call                               | Downloading weights via torch.hub  | Set `TORCH_HOME` to a cached location |\n\n---\n\n**Next**: feed embeddings to a classifier. For structure prediction, use\n`esmfold2`.\n",{"name":5,"license":15,"category":24,"metadata":25,"description":33,"requirements":34},"biomodels",{"third_party":26,"display-name":29},[27],{"kind":28,"name":29,"license":30,"provider":31,"terms_url":32},"weights","ESM-2","MIT","Meta AI","https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fesm\u002Fblob\u002Fmain\u002FLICENSE","Embed proteins with Meta AI's ESM-2 (`fair-esm` package). Use this skill when: (1) Extracting per-residue or per-sequence embeddings for downstream ML, (2) Masked-LM likelihood \u002F mutation effect scoring, (3) Contact prediction from a sequence.\n",[35],"gpu","main","e71245365df78e06cd166602744bf2491f79910e","research-skills-hub\u002Fclaude-science-skills\u002Fskills\u002Ffair-esm2",0,"https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS","https:\u002F\u002Fpengqianhan.github.io\u002FAI-Human-Research-OS\u002F","https:\u002F\u002Fgithub.com\u002Fpengqianhan","Ph.D. candidate of The University of Auckland, New Zealand","2026-11-14 18:08:16",{"source":8,"skillName":5,"commands":46,"prompt":56,"zipAvailable":57},[47,50,53],{"tool":48,"command":49},"npx","npx skills add pengqianhan\u002FAI-Human-Research-OS --skill fair-esm2",{"tool":51,"command":52},"bunx","bunx skills add pengqianhan\u002FAI-Human-Research-OS --skill fair-esm2",{"tool":54,"command":55},"pnpm","pnpm dlx skills add pengqianhan\u002FAI-Human-Research-OS --skill fair-esm2","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fclaude-science-skills\u002Fskills\u002Ffair-esm2\u002FSKILL.md",true,[],[60,67,74,81,88,95],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":30,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M74MWYTHT498SVMN823896T7","workflow-skill-creator","Distills a completed user workflow or interaction into a reusable agent skill. Use when the user asks to turn their workflow, interaction, or multi-step process into a skill, or when they say \"make this a skill\", \"create a skill from what we just did\", \"package this workflow\" or similar. Do not use for creating skills from scratch without an existing workflow (use a generic skill-creator for that).","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Fworkflow_skill_creator\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":30,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M74MWWVCFZ8HQS8840F1V1FT","uv","Checks whether the uv Python package manager is installed and installs it if missing. Ensures uv is on PATH. Use when another skill requires uv as a prerequisite.","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Fuv\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":30,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M74MWRH5BTSWZ2C5PCTD5FV9","uniprot-database","Access protein metadata, function, taxonomy, and sequences across UniProtKB, UniParc, and UniRef. Use when searching for proteins, mapping identifiers, or retrieving functional annotations and publications. Don't use for sequence alignment, protein folding, or sequence similarity search (use specialized skills for those tasks).","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Funiprot_database\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":30,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M74MWN1D3ER9F1FZPWH8HQC1","unibind-database","Queries the UniBind database for experimentally validated transcription factor (TF) binding sites. Use when retrieving direct TF-DNA interaction datasets, downloading binding site coordinates (BED\u002FFASTA) for local analysis, or listing available datasets by species, cell line, or TF name. Don't use to query specific intervals, locations, genes, motif models or expression data.","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Funibind_database\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":30,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":19},"01M74MWHHB5QZY3NR2FYA1EK4N","ucsc-conservation-and-tfbs","Fetch Evolutionary Conservation scores (phyloP, phastCons) and Transcription Factor Binding Sites (TFBS) from the UCSC Genome Browser. Use when analyzing whether genomic variants or regions are evolutionarily conserved, functionally important, or bounded by TF regulators across major projects (ENCODE, JASPAR, ReMap).","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Fucsc_conservation_and_tfbs\u002FSKILL.md",[],[],{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":30,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":19},"01M74MWERYHSY87P21FDMVE6BA","string-database","Query the STRING database for protein-protein interactions (PPIs), functional enrichment, and homology. Use when the user asks about interactions between specific proteins, interaction evidence, confidence scores, protein interaction partners, or pathway enrichments.","https:\u002F\u002Fgithub.com\u002Fpengqianhan\u002FAI-Human-Research-OS\u002Fblob\u002Fmain\u002Fresearch-skills-hub\u002Fscience-skills\u002Fstring_database\u002FSKILL.md",[],[]]