#454 by GitHub stars
HughYau/AcademicForge
One Forge, All Skills: A curated skill collection for academic writing and research. 点开即用，按需配置的一站式学术研究skills平台。
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
skills/claude-science/algorithmic-art/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/alphafold2/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/boltz/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/borzoi/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/chai1/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/compute-env-setup/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/customize/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/diffdock/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/esmfold2/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/evo2/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/fair-esm2/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/figure-composer/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/figure-style/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/indication-dossier/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/learn/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/ligandmpnn/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/literature-review/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/managed-model-endpoints/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/openfold3/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/paper-narrative/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/pdf-explore/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/product-self-knowledge/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/proteinmpnn/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/remote-compute-modal/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/remote-compute-ssh/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/scgpt/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/scvi-tools/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/self-awareness/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/skill-creator/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/solublempnn/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/using-model-endpoint/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/claude-science/web-artifacts-builder/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
skills/scientific-visualization/SKILL.md
Week of Oct 26, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
01b6d90c5b50 → 01b6d90c5b50
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
2.6kGitHub stars
14stars in 7 days
33skills in this repository
1commit in 30 days
Heat
5.5- Star growth, last 7 daysAhead of 73% of tracked repositories
- Star growth, last 24 hoursAhead of 82% of tracked repositories
- Rank movement, last 7 daysAhead of 28% of tracked repositories
- Commits, last 30 daysAhead of 54% of tracked repositories
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekAhead of 0% of tracked repositories
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
Individual skills in this repository
- algorithmic-art
skills/claude-science/algorithmic-art - alphafold2
skills/claude-science/alphafold2 - boltz
skills/claude-science/boltz - borzoi
skills/claude-science/borzoi - chai1
skills/claude-science/chai1 - compute-env-setup
skills/claude-science/compute-env-setup - customize
skills/claude-science/customize - diffdock
skills/claude-science/diffdock - esmfold2
skills/claude-science/esmfold2 - evo2
skills/claude-science/evo2 - fair-esm2
skills/claude-science/fair-esm2 - figure-composer
skills/claude-science/figure-composer - figure-style
skills/claude-science/figure-style - indication-dossier
skills/claude-science/indication-dossier - learn
skills/claude-science/learn - ligandmpnn
skills/claude-science/ligandmpnn - literature-review
skills/claude-science/literature-review - managed-model-endpoints
skills/claude-science/managed-model-endpoints - openfold3
skills/claude-science/openfold3 - paper-narrative
skills/claude-science/paper-narrative - pdf-explore
skills/claude-science/pdf-explore - product-self-knowledge
skills/claude-science/product-self-knowledge - proteinmpnn
skills/claude-science/proteinmpnn - remote-compute-modal
skills/claude-science/remote-compute-modal - remote-compute-ssh
skills/claude-science/remote-compute-ssh - scgpt
skills/claude-science/scgpt - scvi-tools
skills/claude-science/scvi-tools - self-awareness
skills/claude-science/self-awareness - skill-creator
skills/claude-science/skill-creator - solublempnn
skills/claude-science/solublempnn - using-model-endpoint
skills/claude-science/using-model-endpoint - web-artifacts-builder
skills/claude-science/web-artifacts-builder - scientific-visualization
skills/scientific-visualization
How it works
algorithmic-art
When your agent uses it
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
What it loads into context
Frontmatter
73 tokensAlways available to the agent
SKILL.md
4,180 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- ALGORITHMIC PHILOSOPHY CREATION
- DEDUCING THE CONCEPTUAL SEED
- P5.JS IMPLEMENTATION
- INTERACTIVE ARTIFACT CREATION
- VARIATIONS & EXPLORATION
- THE CREATIVE PROCESS
- RESOURCES
Scripts executed separately
templates/generator_template.js
Files
.catalog_stampasset27 bytesLICENSE.txtdoc11,357 bytesSKILL.mdcore20,173 bytestemplates/generator_template.jsruns8,048 bytestemplates/viewer.htmlasset21,442 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubalphafold2
When your agent uses it
Predict protein structure for monomers and multimers with AlphaFold2 via the ColabFold runner (Mirdita et al. 2022, github.com/sokrypton/ColabFold; AlphaFold2 Jumper et al. 2021). Reach for this skill to fold a sequence or complex with the AF2/AF2-Multimer evoformer, to validate designed sequences by self-consistency pLDDT, ipTM, and RMSD, or to run a quick MSA-backed prediction using the public MMseqs2 server.
What it loads into context
Frontmatter
385 tokensAlways available to the agent
SKILL.md
1,381 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Running it
- Unified-memory defaults loop forever under gVisor — the env patches them out
- The MSA server is the wall-clock bottleneck, and it is shared
- Errors worth recognizing
Files
.catalog_stampasset13 bytesSKILL.mdcore4,965 bytesSKILL.mdcore4,965 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubboltz
When your agent uses it
Structure prediction for protein, nucleic-acid, and small-molecule complexes with Boltz-2 (Passaro & Wohlwend et al. 2025, github.com/jwohlwend/boltz). Reach for this skill to validate designed binders against a target, to co-fold a protein with a SMILES or CCD ligand, or to get an open-source AlphaFold3 alternative with optional binding-affinity prediction.
What it loads into context
Frontmatter
310 tokensAlways available to the agent
SKILL.md
1,475 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Running it
- Affinity head
- `msa: empty` is an accuracy hit, not a memory save
- Missing fast kernels are slow, not fatal
- Errors worth recognizing
Files
.catalog_stampasset13 bytesSKILL.mdcore5,305 bytesSKILL.mdcore5,305 bytesSKILL.mdcore5,305 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubborzoi
When your agent uses it
Predict genome-wide functional tracks (RNA-seq, CAGE, DNase, ChIP) from DNA sequence with Borzoi. Use this skill when: (1) Scoring the regulatory effect of a variant on expression/accessibility, (2) Generating predicted coverage tracks for a locus, (3) Prioritising non-coding variants by predicted track delta.
What it loads into context
Frontmatter
302 tokensAlways available to the agent
SKILL.md
1,077 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Prerequisites
- How to run
- Output format
- Remote compute
- Troubleshooting
Files
.catalog_stampasset13 bytesSKILL.mdcore3,916 bytesSKILL.mdcore3,916 bytesSKILL.mdcore3,916 bytesSKILL.mdcore3,916 bytesSKILL.mdcore3,916 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubchai1
When your agent uses it
Structure prediction for protein, nucleic-acid, and small-molecule complexes with the Chai-1 foundation model (Chai Discovery 2024, github.com/chaidiscovery/chai-lab). Reach for this skill to predict an antibody-antigen or protein-ligand complex from a single FASTA, to re-fold designed binders as an AlphaFold-multimer alternative, or to drive co-folding from Python for batched campaigns on a GPU.
What it loads into context
Frontmatter
342 tokensAlways available to the agent
SKILL.md
1,280 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Running it
- Unset `CHAI_DOWNLOADS_DIR` fails mid-run with PermissionError on a read-only image
- No-MSA mode still loads a 3 B-parameter ESM — same VRAM, not less
- Errors worth recognizing
Files
.catalog_stampasset13 bytesSKILL.mdcore4,615 bytesSKILL.mdcore4,615 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubcompute-env-setup
When your agent uses it
Set up a compute environment on a remote provider so Claude Science jobs can run there. Covers direct SSH/conda hosts, Slurm clusters, container-via-bridge runners, and managed-API providers (Modal, GCP, RunPod). Use when standing up a new provider, porting an env to a different backend, adding a tool that needs its own software stack, or wiring weight caches. Triggers on "new compute provider", "set up env on", "port env to", "build GPU image", "weight cache", "compute_details", "conda env on the box", "apptainer on slurm".
What it loads into context
Frontmatter
145 tokensAlways available to the agent
SKILL.md
4,189 tokensWhen the skill is activated
Reference documents
5,750 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- Provider shapes
- The declarative spec
- BYOC providers — building via the `compute_provider` kernel
- Weights
- Validation
- Diagnosing failures
- compute_details — recording what's set up
- Reference guides
Files
.catalog_stampasset13 bytesSKILL.mdcore17,812 bytesSKILL.mdcore17,812 bytesSKILL.mdcore17,812 bytesSKILL.mdcore17,812 bytesSKILL.mdcore17,812 bytesreferences/envs_reference.mddoc18,249 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubcustomize
When your agent uses it
Create, configure, and maintain custom agent profiles and author new skills via the `repl` tool. Use when the user wants to create an agent profile, build a custom agent, modify agent capabilities, attach or detach skills/connectors on a profile, author a skill, or inspect which connectors and tools are available. Also use whenever you need the `host.agents.*` or `host.skills.*` Python SDK.
What it loads into context
Frontmatter
98 tokensAlways available to the agent
SKILL.md
5,336 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Python SDK
- Workflow: scope → draft → review → create
- Authoring skills
- Kernel sidecars (`kernel.py` / `kernel.R`)
Files
.catalog_stampasset13 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytesSKILL.mdcore21,748 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubdiffdock
When your agent uses it
Predict small-molecule binding poses with DiffDock-L (Corso et al. 2023/2024, github.com/gcorso/DiffDock) — blind diffusion docking that places a ligand into a protein pocket without a predefined search box and ranks the samples with a learned confidence model. Reach for this skill to dock a SMILES or SDF against a PDB, to generate ranked 3D poses for a small fragment library, or to get a starting pose for downstream rescoring. DiffDock predicts geometry, not affinity.
What it loads into context
Frontmatter
233 tokensAlways available to the agent
SKILL.md
1,164 tokensWhen the skill is activated
Reference documents
531 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Running it
- The YAML config overwrites your CLI flags
- The first run is silent for ~11 minutes and needs ≥32 GB host RAM
- The README's `--ligand` works on the CLI by accident — use `--ligand_description`
- Errors worth recognizing
Files
.catalog_stampasset13 bytesSKILL.mdcore4,476 bytesSKILL.mdcore4,476 bytesreferences/workflows.mddoc1,984 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubesmfold2
When your agent uses it
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.
What it loads into context
Frontmatter
397 tokensAlways available to the agent
SKILL.md
3,099 tokensWhen the skill is activated
Reference documents
1,371 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Install
- Usage — local model
- Model variants on HF `biohub/`
- Throughput: `set_kernel_backend("fused")` is REQUIRED
- ESMFold2-Experimental* — design hook
- Gotcha: cusolver SVD poison + structseq constructor
- With-MSA mode
- Paper-matched inference configuration
- Training data cutoff
- ESMC language model
Files
.catalog_stampasset13 bytesSKILL.mdcore10,208 bytesSKILL.mdcore10,208 bytesSKILL.mdcore10,208 bytesSKILL.mdcore10,208 bytesSKILL.mdcore10,208 bytesSKILL.mdcore10,208 bytesSKILL.mdcore10,208 bytesSKILL.mdcore10,208 bytesreferences/design-hook.mddoc1,452 bytesreferences/esmc.mddoc3,052 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubevo2
When your agent uses it
Score, embed, and generate DNA sequences with Evo 2, a long-context genomic foundation model. Use this skill when: (1) Computing per-nucleotide or per-sequence likelihoods for variant effect scoring, (2) Embedding genomic windows for downstream classification, (3) Generating DNA conditioned on a prefix, (4) Scoring regulatory or coding regions across species.
What it loads into context
Frontmatter
238 tokensAlways available to the agent
SKILL.md
1,482 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Prerequisites
- How to run
- Models
- Output format
- Decision tree
- Remote compute
- Typical performance
- Troubleshooting
Files
.catalog_stampasset13 bytesSKILL.mdcore5,418 bytesSKILL.mdcore5,418 bytesSKILL.mdcore5,418 bytesSKILL.mdcore5,418 bytesSKILL.mdcore5,418 bytesSKILL.mdcore5,418 bytesSKILL.mdcore5,418 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubfair-esm2
When your agent uses it
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 / mutation effect scoring, (3) Contact prediction from a sequence.
What it loads into context
Frontmatter
186 tokensAlways available to the agent
SKILL.md
1,393 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Prerequisites
- How to run
- Models
- Output format
- Remote compute
- Troubleshooting
Files
.catalog_stampasset13 bytesSKILL.mdcore4,914 bytesSKILL.mdcore4,914 bytesSKILL.mdcore4,914 bytesSKILL.mdcore4,914 bytesSKILL.mdcore4,914 bytesSKILL.mdcore4,914 bytesSKILL.mdcore4,914 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubfigure-composer
When your agent uses it
Compose one publication-grade multi-panel figure. Entry from a one-line claim + data files, OR from an existing figure via `derive_outline_prompt` (you read the PNG). Runs a per-figure loop: outline (12-col grid, per-panel ask + label_budget) → render each panel with `panel_task` (loading `figure-style`), one at a time or parallelized → tile + stamp letters with `compose_figure` → adversarial composite self-review with two-tier feedback (Tier-1 outline_revisions / Tier-2 per-panel violations) → regen affected panels, ≤3 rounds. Helpers: panel_task / compose_figure / compose_crops / composite_review_task / derive_outline_prompt. For one standalone plot use `figure-style`; for whole-paper figure ordering use `paper-narrative`.
What it loads into context
Frontmatter
182 tokensAlways available to the agent
SKILL.md
2,501 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Setup (any agent, no API key)
- Inputs
- 0. Where this sits
- Entry points (pick one)
- 1. Narrative → panel outline
- 2. Render the panels (one at a time, or parallel)
- 3. Compose
- 3.5 Look before you review (vision self-QA)
- 4. Adversarial self-review loop (two-tier, design rules held fixed)
- Anti-patterns
Scripts executed separately
kernel.py
Files
.catalog_stampasset13 bytesSKILL.mdcore9,916 bytesSKILL.mdcore9,916 bytesSKILL.mdcore9,916 bytesSKILL.mdcore9,916 byteskernel.pyruns14,050 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubfigure-style
When your agent uses it
Publication-grade figure correctness and legibility rules. Load before drawing any plot and call `apply_figure_style()` — sets a role-mapped font-size ladder, outward ticks, frameless legends, and 300-dpi output. The skill is a checklist, not a house look: data fidelity (claim-titles tested against every row, excluded data never enters summaries), label economy (floor and ceiling), colour threading, chart-choice-by-data-shape, layout, and a render-then-verify QA loop (bbox collision + per-panel perceptual check). Ships helpers: focal_palette, bar_with_points, strip_with_median, end_of_line_labels, panel_letter, set_frame, panel_crops. For multi-panel figures load `figure-composer`; for whole-paper figure arc load `paper-narrative`.
What it loads into context
Frontmatter
180 tokensAlways available to the agent
SKILL.md
4,034 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- §0 Scope
- §1 Data fidelity & self-consistency
- §2 Label economy — floor and ceiling
- §3 Axes, scales, small multiples
- §4 Colour
- §5 Typography
- §6 Chart-family guidance (by data shape)
- §7 Layout & narrative
- §8 Anti-patterns
- §9 Render-then-verify
Scripts executed separately
kernel.py
Files
.catalog_stampasset13 bytesSKILL.mdcore16,728 bytesSKILL.mdcore16,728 bytesSKILL.mdcore16,728 bytesSKILL.mdcore16,728 bytesSKILL.mdcore16,728 byteskernel.pyruns13,428 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubindication-dossier
When your agent uses it
Generate a therapeutic indication dossier. Covers the patient population, epidemiology, disease biology, standard of care, regulatory precedent, and landmark clinical trials.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
1,357 tokensWhen the skill is activated
Reference documents
5,196 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Framing
- Inputs
- Tools this skill expects
- Output layout
- Before starting
- Workflow
- Resuming
Files
.catalog_stampasset13 bytesSKILL.mdcore5,875 bytesreferences/00-research-standards.mddoc3,101 bytesreferences/01-meta-initialization.mddoc1,724 bytesreferences/02-epidemiology-research.mddoc2,109 bytesreferences/03-biology-soc-research.mddoc2,368 bytesreferences/04-regulatory-trials-research.mddoc3,154 bytesreferences/05-synthesis.mddoc5,323 bytesreferences/06-writing-style.mddoc1,746 bytesreferences/waypoint-schemas.mddoc2,923 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHublearn
When your agent uses it
Use this skill when the user wants intellectual understanding — learning how or why something works, not getting a task done or soliciting Claude's judgment. Trigger for: - Explicit learning requests: teach, explain, ELI5, walk me through, quiz me, flashcards, "I'm rusty on"; definitions ("what is X") - Terse concept names implying "help me understand this": "Galois theory," "transformers, from scratch" - Confusion signals: "won't stick," "keep mixing these up," "not getting it" - Learning-path questions: prerequisites, sequencing, what to study before X - Conceptual questions about mechanisms, causes, or dynamics Don't trigger for: - Tasks: coding, writing, calculation, translation, factual lookup, news updates - Personal troubleshooting; resource/textbook recommendations - Claude's evaluative verdict: opinion prompts ("do you think X", "settle this", "honest take", "is X dead / still taken seriously") and interpretive takes ("was X really as harsh as people say")
What it loads into context
Frontmatter
243 tokensAlways available to the agent
SKILL.md
2,801 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Diagnose before you teach
- The core rhythm: one step forward, every turn
- Holding the line under pressure
- A toolkit of moves
- Showing, not just telling
- Academic integrity — when it applies
- What consistently goes wrong
- Tone
Files
.catalog_stampasset27 bytesLICENSE.txtdoc11,357 bytesSKILL.mdcore13,448 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubligandmpnn
When your agent uses it
Inverse-fold a backbone with ligand, nucleic-acid, and metal context using LigandMPNN (Dauparas et al. 2023, github.com/dauparas/LigandMPNN). Reach for this skill to redesign the residues lining a binding pocket around a bound small molecule or cofactor, to design metal-coordinating sites where the geometry must be respected, or to get threaded designed-sequence PDBs out of any MPNN run.
What it loads into context
Frontmatter
222 tokensAlways available to the agent
SKILL.md
1,657 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Running it
- Model types — which one to pick
- ProDy compiles from source on py3.11 — `pip install` fails without a C compiler
- Turning ligand context off changes the answer, not the model
- Stripped HETATM or a chain filter silently drops the ligand — the design comes back pocket-blind
- Errors worth recognizing
Files
.catalog_stampasset13 bytesSKILL.mdcore6,217 bytesSKILL.mdcore6,217 bytesSKILL.mdcore6,217 bytesSKILL.mdcore6,217 bytesSKILL.mdcore6,217 bytesSKILL.mdcore6,217 bytesSKILL.mdcore6,217 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubliterature-review
When your agent uses it
Find, verify, and synthesize scientific literature — from "what's the seminal paper for X" through full multi-source reviews. Covers grounding claims in real retrieved sources, avoiding fabricated citations, handling retractions, and calibrating confidence to evidence strength.
What it loads into context
Frontmatter
299 tokensAlways available to the agent
SKILL.md
2,704 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Setup (any agent, no API key)
- Read the request for what it's actually asking
- Grounding: retrieve first, then write
- Retractions and the null result
- Synthesis is comparison, not summary
- Making the prose carry its weight
- Write prose, not a bulleted bibliography
- Calibrating to evidence
- Put the answer in the answer — and open on the substance
- Style pass before saving
Scripts executed separately
kernel.py
Files
.catalog_stampasset13 bytesSKILL.mdcore12,239 bytesSKILL.mdcore12,239 bytesSKILL.mdcore12,239 byteskernel.pyruns21,422 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubmanaged-model-endpoints
When your agent uses it
Register a model service in the managed family — a local model server container the daemon starts/stops on demand, or a remote upstream model API (https). Read the runbook, allocate a port (local only), compose idempotent start/stop scripts (local only), register once. Load when the user wants a model service available for inference, or when list_compute shows managed endpoints.
What it loads into context
Frontmatter
94 tokensAlways available to the agent
SKILL.md
2,617 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Calling a registered endpoint — inference cells
- Enablement — once per machine
- Register (repl kernel)
- Remote endpoints — upstream APIs (no lifecycle)
- Composing the start script
- Failures
Files
.catalog_stampasset13 bytesSKILL.mdcore10,718 bytesSKILL.mdcore10,718 bytesSKILL.mdcore10,718 bytesSKILL.mdcore10,718 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubopenfold3
When your agent uses it
Structure prediction using OpenFold3, an open-weights PyTorch reproduction of AlphaFold3 from the AlQuraishi Lab. Use this skill when predicting protein/nucleic-acid/ligand complex structures with an Apache-2.0-licensed AF3 reimplementation.
What it loads into context
Frontmatter
297 tokensAlways available to the agent
SKILL.md
2,150 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Prerequisites
- How to run
- Query JSON format
- Key parameters
- Output format
- What good output looks like
- Verify
- Troubleshooting
Files
.catalog_stampasset13 bytesSKILL.mdcore7,378 bytesSKILL.mdcore7,378 bytesSKILL.mdcore7,378 bytesSKILL.mdcore7,378 bytesSKILL.mdcore7,378 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubpaper-narrative
When your agent uses it
Judge and reshape the STORY a paper's figures tell. Input is the work itself — manuscript (or abstract) + figure deck — no hand-written brief. `paper_brief_prompt(abstract, captions)` hands you the prompt to write the brief yourself (pitch/vision/per-figure-claims); then you play a handling editor over the full deck and return hook_verdict (would Fig 1 make me send this for review?), arc (hook→mechanism→evidence→application), figure_moves (panels in the wrong figure), missing_panels (concrete analyses to RUN), kill_list, and boldest_defensible_fig1. Hands per-figure claims to `figure-composer`. Load when writing or revising a paper.
What it loads into context
Frontmatter
167 tokensAlways available to the agent
SKILL.md
951 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Setup (any agent, no API key)
- When to load
- Workflow
- Minimal invocation
Scripts executed separately
kernel.py
Files
.catalog_stampasset13 bytesSKILL.mdcore3,781 bytesSKILL.mdcore3,781 byteskernel.pyruns6,213 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubpdf-explore
When your agent uses it
Use this skill when the user has attached a PDF, paper, report, or other document and the answer needs content from more than one place in it: summarize the methods or any other section, compare sections, find where a topic is discussed, read a value or label off a figure or chart, or find/list/extract every instance of something across the whole document (datasets, benchmarks, citations, figures, table rows, accession numbers — including appendices). Parses the PDF once with a deterministic Python kernel: `pdf_pages` (pages as persistent text, or high-res images), `pdf_outline` (embedded TOC), `pdf_scan` (a lexical pre-filter that narrows a long doc to candidate pages), `pdf_grep` (regex sweep for exhaustive pattern extraction). You read the shortlist and do the relevance / summary / extraction judgment yourself. For PDF creation/manipulation, use reportlab/pypdf directly.
What it loads into context
Frontmatter
204 tokensAlways available to the agent
SKILL.md
3,257 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Setup (any agent, no API key)
- Which helper
- Recipe — pull the sections you need as persistent text (synthesis)
- Recipe — navigate by outline (try this first)
- Recipe — find the pages relevant to a query
- Recipe — read a figure in detail
- Recipe — list / extract every instance of X across the doc
- When NOT to use this skill
- Mode (scanned PDFs)
- Cost & budget
- Caching
Scripts executed separately
kernel.py
Files
.catalog_stampasset13 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 bytesSKILL.mdcore12,424 byteskernel.pyruns19,642 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubproduct-self-knowledge
When your agent uses it
Stop and consult this skill whenever your response would include specific facts about Anthropic's products. Covers: Claude Code (how to install, Node.js requirements, platform/OS support, MCP server integration, configuration), Claude API (function calling/tool use, batch processing, SDK usage, rate limits, pricing, models, streaming), and Claude.ai (Pro vs Team vs Enterprise plans, feature limits). Trigger this even for coding tasks that use the Anthropic SDK, content creation mentioning Claude capabilities or pricing, or LLM provider comparisons. Any time you would otherwise rely on memory for Anthropic product details, verify here instead — your training data may be outdated or wrong.
What it loads into context
Frontmatter
150 tokensAlways available to the agent
SKILL.md
590 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Core Principles
- Question Routing
- Response Workflow
- Quick Reference
Files
.catalog_stampasset13 bytesSKILL.mdcore2,537 bytesSKILL.mdcore2,537 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubproteinmpnn
When your agent uses it
Inverse-fold a protein backbone (PDB structure) into amino-acid sequence with ProteinMPNN (Dauparas et al. 2022, github.com/dauparas/ProteinMPNN). Reach for this skill to run sequence design on RFdiffusion backbones, to redesign one chain of a PDB while holding interface residues fixed, or to generate a temperature-swept set of sequences for downstream folding.
What it loads into context
Frontmatter
205 tokensAlways available to the agent
SKILL.md
1,242 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Running it
- A flat chain map in `--fixed_positions_jsonl` silently redesigns every residue
- Checkpoints — which one to pick
- Errors worth recognizing
Files
.catalog_stampasset13 bytesSKILL.mdcore4,587 bytesSKILL.mdcore4,587 bytesSKILL.mdcore4,587 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubremote-compute-modal
When your agent uses it
Run GPU jobs on the user's own Modal account via host.compute.create('byoc:modal', ...). Covers the create→submit→wait_for_notification flow, the compute_provider kernel for env setup, image/volume resolution, and the two approval cards. Load once you've decided to dispatch to Modal.
What it loads into context
Frontmatter
76 tokensAlways available to the agent
SKILL.md
10,803 tokensWhen the skill is activated
Reference documents
3,790 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Two timeouts, one timeline
- Two surfaces, one provider
- Workflow
- The image-reference rule
- What latency to expect
- When the user gives you a budget
- Warm reuse, reattachment, and `close()`
- When the job fails
- What to record
- GPU tier reference
- Volumes
- Modal Environment (Settings → Compute → modal → Environment)
- Network egress from the job sandbox
- Common env gotchas
Scripts executed separately
envs/chemistry_gpu.pyenvs/esmfold2_gpu.pyenvs/genomics_evo2_gpu.pyenvs/proteomics_boltz_gpu.pyenvs/proteomics_gpu.pyenvs/proteomics_jax_gpu.pyenvs/proteomics_openfold_gpu.pyenvs/proteomics_rfd_diffdock_gpu.pyenvs/singlecell_gpu.pyprovider.py
Files
.catalog_stampasset13 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesSKILL.mdcore43,964 bytesenv-setup.mddoc14,431 bytesenvs/chemistry_gpu.pyruns1,080 bytesenvs/esmfold2_gpu.pyruns7,122 bytesenvs/genomics_evo2_gpu.pyruns2,968 bytesenvs/proteomics_boltz_gpu.pyruns2,272 bytesenvs/proteomics_gpu.pyruns3,198 bytesenvs/proteomics_jax_gpu.pyruns2,782 bytesenvs/proteomics_openfold_gpu.pyruns5,218 bytesenvs/proteomics_rfd_diffdock_gpu.pyruns5,144 bytesenvs/singlecell_gpu.pyruns4,564 bytesprovider.jsonasset409 bytesprovider.pyruns39,816 bytesrequirements.lockasset72,112 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubremote-compute-ssh
When your agent uses it
Submit→wait_for_notification→harvest workflow for the user's SSH/SLURM hosts. Load once you've decided to dispatch remote.
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
4,791 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Workflow
- What to record
- When the job fails
- `c.submit_job()` on SSH
- When the user gives you a budget
- Submitting several jobs
- When the user asks you to set up the host
- When it's unclear what's available on the host
Files
.catalog_stampasset13 bytesSKILL.mdcore19,671 bytesSKILL.mdcore19,671 bytesSKILL.mdcore19,671 bytesSKILL.mdcore19,671 bytesSKILL.mdcore19,671 bytesSKILL.mdcore19,671 bytesSKILL.mdcore19,671 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubscgpt
When your agent uses it
Embed and annotate single-cell expression data with scGPT, a foundation model for single-cell biology. Use this skill when: (1) Producing cell embeddings from an AnnData for clustering/integration, (2) Zero-shot or fine-tuned cell-type annotation, (3) Gene-level representation for perturbation/GRN tasks. For probabilistic single-cell models (scVI etc.), use the scvi-tools library.
What it loads into context
Frontmatter
260 tokensAlways available to the agent
SKILL.md
1,368 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Prerequisites
- How to run
- Output format
- Remote compute
- Gotchas
- Troubleshooting
Files
.catalog_stampasset13 bytesSKILL.mdcore5,347 bytesSKILL.mdcore5,347 bytesSKILL.mdcore5,347 bytesSKILL.mdcore5,347 bytesSKILL.mdcore5,347 bytesSKILL.mdcore5,347 bytesSKILL.mdcore5,347 bytesSKILL.mdcore5,347 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubscvi-tools
When your agent uses it
Probabilistic single-cell RNA-seq with scvi-tools — scVI for a batch-corrected latent space, scANVI for semi-supervised label transfer, and Bayesian differential expression. Reach for this skill to integrate scRNA-seq batches, embed cells for clustering, transfer annotations from a reference onto a query, or score differentially expressed genes per cluster. For spatial deconvolution / mapping use the cell2location, DestVI, or Tangram methods instead.
What it loads into context
Frontmatter
138 tokensAlways available to the agent
SKILL.md
2,575 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Setup (any agent, no API key)
- How to run
- Differential expression
- Output format
- Remote compute (rent a GPU)
- Gotchas
- Troubleshooting
Scripts executed separately
kernel.py
Files
.catalog_stampasset13 bytesSKILL.mdcore9,368 bytesSKILL.mdcore9,368 bytesSKILL.mdcore9,368 bytesSKILL.mdcore9,368 bytesSKILL.mdcore9,368 bytesSKILL.mdcore9,368 byteskernel.pyruns2,248 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubself-awareness
When your agent uses it
Claude Science's own session database schema and SDK surface for introspection via host.query(). Load this when you need to query your own conversation history, token usage, cost accounting, execution log, or artifact metadata beyond what host.frames()/host.artifacts() provide — e.g. "how many tokens has this session used", "what was my last tool call", "list every file I've written", "where are messages stored", "what tables can I query", "inspect frames.context_data", or any time you're about to PRAGMA-probe the Claude Science metadata DB to discover its schema.
What it loads into context
Frontmatter
132 tokensAlways available to the agent
SKILL.md
3,646 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Dialect and limits
- Queryable tables
- Denied tables
- Worked examples
- SDK surface — which tool runs what
Files
.catalog_stampasset13 bytesSKILL.mdcore13,845 bytesSKILL.mdcore13,845 bytesSKILL.mdcore13,845 bytesSKILL.mdcore13,845 bytesSKILL.mdcore13,845 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubskill-creator
When your agent uses it
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
What it loads into context
Frontmatter
75 tokensAlways available to the agent
SKILL.md
8,455 tokensWhen the skill is activated
Reference documents
9,508 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Reading and writing skill files — `host.skills.*`
- Communicating with the user
- Creating a skill
- Running and evaluating test cases
- Improving the skill
- Advanced: Blind comparison
- Description Optimization
- Claude.ai-specific instructions
- Cowork-Specific Instructions
- Reference files
Scripts executed separately
eval-viewer/generate_review.pyscripts/__init__.pyscripts/aggregate_benchmark.pyscripts/generate_report.pyscripts/improve_description.pyscripts/package_skill.pyscripts/quick_validate.pyscripts/run_eval.pyscripts/run_loop.pyscripts/utils.py
Files
.catalog_stampasset13 bytesSKILL.mdcore38,235 bytesSKILL.mdcore38,235 bytesSKILL.mdcore38,235 bytesSKILL.mdcore38,235 bytesSKILL.mdcore38,235 bytesSKILL.mdcore38,235 bytesSKILL.mdcore38,235 bytesSKILL.mdcore38,235 bytesSKILL.mdcore38,235 bytesagents/analyzer.mddoc10,546 bytesagents/comparator.mddoc7,370 bytesagents/grader.mddoc9,053 bytesassets/eval_review.htmlasset8,502 byteseval-viewer/generate_review.pyruns16,365 byteseval-viewer/viewer.htmlasset50,995 bytesreferences/schemas.mddoc12,088 bytesscripts/__init__.pyruns0 bytesscripts/aggregate_benchmark.pyruns14,500 bytesscripts/generate_report.pyruns12,847 bytesscripts/improve_description.pyruns11,116 bytesscripts/package_skill.pyruns4,198 bytesscripts/quick_validate.pyruns3,984 bytesscripts/run_eval.pyruns11,464 bytesscripts/run_loop.pyruns13,605 bytesscripts/utils.pyruns1,661 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubsolublempnn
When your agent uses it
Inverse-fold a backbone with SolubleMPNN — ProteinMPNN retrained on a soluble-PDB subset (Dauparas et al. 2022) — for sequences biased toward cytosolic expression and reduced aggregation. Reach for this skill when designs from vanilla ProteinMPNN are aggregating or going to inclusion bodies, when redesigning a membrane-adjacent fold for soluble expression, or when an E. coli expression screen is the next step.
What it loads into context
Frontmatter
232 tokensAlways available to the agent
SKILL.md
1,072 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Running it
- Hydrophobic surface patches still recur where the fold needs them
- "Crystallisable" training set ≠ "soluble in your host" — keep an orthogonal filter
Files
.catalog_stampasset13 bytesSKILL.mdcore4,277 bytesSKILL.mdcore4,277 bytesSKILL.mdcore4,277 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubusing-model-endpoint
When your agent uses it
Call a registered model endpoint over its native HTTP API from the endpoint's scoped inference kernel (BASE_URL preloaded). Load once a task needs predictions from a registered model endpoint.
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
421 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Scripts executed separately
provider.py
Files
.catalog_stampasset13 bytesSKILL.mdcore1,754 bytesprovider.jsonasset243 bytesprovider.pyruns5,246 bytesrequirements.lockasset1,580 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubweb-artifacts-builder
When your agent uses it
Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.
What it loads into context
Frontmatter
75 tokensAlways available to the agent
SKILL.md
699 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Design & Style Guidelines
- Quick Start
- Reference
Scripts executed separately
scripts/bundle-artifact.shscripts/init-artifact.sh
Files
.catalog_stampasset27 bytesLICENSE.txtdoc11,357 bytesSKILL.mdcore3,087 bytesSKILL.mdcore3,087 bytesSKILL.mdcore3,087 bytesSKILL.mdcore3,087 bytesscripts/bundle-artifact.shruns1,517 bytesscripts/init-artifact.shruns9,924 bytesscripts/shadcn-components.tar.gzasset19,967 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubscientific-visualization
When your agent uses it
Create publication figures with matplotlib/seaborn/plotly. Multi-panel layouts, error bars, significance markers, colorblind-safe, export PDF/EPS/TIFF, for journal-ready scientific plots.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
6,444 tokensWhen the skill is activated
Reference documents
12,797 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Overview
- When to Use This Skill
- Quick Start Guide
- Core Principles and Best Practices
- Common Tasks
- Statistical Rigor
- Working with Different Plotting Libraries
- Resources
- Workflow Summary
- Common Pitfalls to Avoid
- Final Checklist
Scripts executed separately
assets/color_palettes.pyscripts/figure_export.pyscripts/style_presets.py
Files
SKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesSKILL.mdcore25,298 bytesassets/color_palettes.pyruns5,754 bytesassets/nature.mplstyleasset1,449 bytesassets/presentation.mplstyleasset1,309 bytesassets/publication.mplstyleasset1,474 bytesreferences/color_palettes.mddoc9,861 bytesreferences/journal_requirements.mddoc9,533 bytesreferences/matplotlib_examples.mddoc18,103 bytesreferences/publication_guidelines.mddoc8,556 bytesscripts/figure_export.pyruns10,904 bytesscripts/style_presets.pyruns12,199 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubSkills included
- algorithmic-art
- alphafold2
- boltz
- borzoi
- chai1
- compute-env-setup
- customize
- diffdock
- esmfold2
- evo2
- fair-esm2
- figure-composer
- figure-style
- indication-dossier
- learn
- ligandmpnn
- literature-review
- managed-model-endpoints
- openfold3
- paper-narrative
- pdf-explore
- product-self-knowledge
- proteinmpnn
- remote-compute-modal
- remote-compute-ssh
- scgpt
- scvi-tools
- self-awareness
- skill-creator
- solublempnn
- using-model-endpoint
- web-artifacts-builder
- scientific-visualization
Repository facts
- License
- No license detected
- Created
- Feb 3, 2026
- Last pushed
- Nov 1, 2026
- Forks
- 155
- Open issues
- 2
- Project activity
- Active
Rank over the last 12 weeks
Now #454 · best #331 · down 73 in 30 days
- #443 on Nov 1
- #447 on Nov 2
- #449 on Nov 3
- #450 on Nov 4
- #452 on Nov 5
- #454 since Nov 6
74 earlier periods
- #320 on Aug 2
- #322 on Aug 7
- #326 on Aug 8
- #325 on Aug 9
- #335 from Aug 10 to Aug 11
- #329 on Aug 12
- #333 on Aug 13
- #334 on Aug 14
- #331 on Aug 15
- #335 on Aug 16
- #334 on Aug 17
- #341 on Aug 18
- #340 on Aug 19
- #339 on Aug 20
- #341 on Aug 21
- #336 on Aug 22
- #347 on Aug 24
- #348 on Aug 25
- #347 on Aug 26
- #348 from Aug 27 to Aug 28
- #346 on Aug 29
- #341 on Aug 30
- #350 on Aug 31
- #356 on Sep 1
- #351 on Sep 2
- #358 on Sep 3
- #351 on Sep 4
- #356 on Sep 5
- #354 on Sep 6
- #363 on Sep 7
- #359 on Sep 8
- #358 on Sep 11
- #362 on Sep 12
- #361 on Sep 13
- #365 on Sep 14
- #368 on Sep 15
- #370 on Sep 16
- #367 on Sep 17
- #369 on Sep 18
- #363 on Sep 19
- #374 on Sep 20
- #372 on Sep 21
- #376 on Sep 22
- #370 on Sep 23
- #369 on Sep 24
- #368 on Sep 25
- #364 on Sep 26
- #369 on Sep 27
- #373 on Sep 28
- #381 on Sep 29
- #379 on Sep 30
- #378 on Oct 1
- #380 on Oct 2
- #367 on Oct 4
- #375 from Oct 5 to Oct 6
- #381 on Oct 7
- #387 on Oct 10
- #389 on Oct 11
- #398 on Oct 12
- #396 on Oct 13
- #401 on Oct 14
- #400 on Oct 15
- #404 on Oct 16
- #397 on Oct 17
- #402 on Oct 18
- #407 on Oct 19
- #411 on Oct 20
- #415 on Oct 21
- #414 on Oct 22
- #410 on Oct 24
- #430 on Oct 25
- #438 on Oct 28
- #440 on Oct 29
- #442 from Oct 30 to Oct 31
Before you install
Network or broad shellAutomated signals, not a guarantee
Checked Nov 2, 2026
Changed on Oct 27, 2026
No longer found: Access to secrets
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 172 notes
skills/claude-science/alphafold2/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/boltz/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/boltz/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/borzoi/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/borzoi/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/borzoi/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/borzoi/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/chai1/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/compute-env-setup/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/compute-env-setup/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/compute-env-setup/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/compute-env-setup/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/customize/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/customize/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/customize/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/customize/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/customize/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/customize/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/customize/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claude-science/customize/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- 20 dangerous shell commandsin 16 files: remote-compute-modal/provider.py:86 managed-model-endpoints/SKILL.md:139 remote-compute-modal/env-setup.md:44 and 13 more
- 302 network referencesin 43 files: literature-review/kernel.py:14 compute-env-setup/references/envs_reference.md:63 remote-compute-modal/envs/proteomics_rfd_diffdock_gpu.py:47 and 40 more
- No hidden or obfuscated text
- 2 instruction override signalsin 1 file: customize/SKILL.md:208
- 33 executable scriptsin 33 files: algorithmic-art/templates/generator_template.js:1 figure-composer/kernel.py:1 figure-style/kernel.py:1 and 30 more
- 44 references to secretsin 12 files: literature-review/kernel.py:14 managed-model-endpoints/SKILL.md:42 remote-compute-modal/provider.py:53 and 9 more
- No requests to vote in Hot
Show all 401 signals by file
skills/claude-science/algorithmic-art/LICENSE.txt
skills/claude-science/algorithmic-art/SKILL.md
- Network accessline 278
<head> <!-- p5.js from CDN - always available --> <script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.7.0/p5.min.js"></script> - Network accessline 279
<!-- p5.js from CDN - always available --> <script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.7.0/p5.min.js"></script> <style> - Network accessline 280
<script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.7.0/p5.min.js"></script>
skills/claude-science/algorithmic-art/templates/generator_template.js
- Executable scriptline 1
Executable file
skills/claude-science/algorithmic-art/templates/viewer.html
- Network accessline 21
<meta name="viewport" content="width=device-width, initial-scale=1.0"> <title>Generative Art Viewer</title> <script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.7.0/p5.min.js"></script> - Network accessline 22
<title>Generative Art Viewer</title> <script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.7.0/p5.min.js"></script> <link rel="preconnect" href="https://fonts.googleapis.com"> - Network accessline 23
<script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.7.0/p5.min.js"></script> - Network accessline 24
<link rel="preconnect" href="https://fonts.googleapis.com"> - Network accessline 25
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin> - Network accessline 26
<link href="https://fonts.googleapis.com/css2?family=Poppins:wght@400;500;600&family=Lora:wght@400;500&display=swap" rel="stylesheet">
skills/claude-science/alphafold2/SKILL.md
- Network accessline 21
provider: Google DeepMind license: CC-BY-4.0 terms_url: https://github.com/google-deepmind/alphafold#model-parameters-license - Network accessline 22
license: CC-BY-4.0 terms_url: https://github.com/google-deepmind/alphafold#model-parameters-license # api.colabfold.com has no published ToS or privacy policy. The GitHub - Network accessline 23
terms_url: https://github.com/google-deepmind/alphafold#model-parameters-license - Network accessline 28
name: ColabFold MSA server (api.colabfold.com) provider: Steinegger Lab info_url: https://github.com/sokrypton/ColabFold/wiki - Network accessline 29
provider: Steinegger Lab info_url: https://github.com/sokrypton/ColabFold/wiki --- - Network accessline 30
info_url: https://github.com/sokrypton/ColabFold/wiki
skills/claude-science/boltz/SKILL.md
- Network accessline 17
name: Boltz-2 license: MIT terms_url: https://github.com/jwohlwend/boltz/blob/main/LICENSE - Network accessline 18
license: MIT terms_url: https://github.com/jwohlwend/boltz/blob/main/LICENSE # SKILL.md — `--use_msa_server` queries api.colabfold.com; the doc says to - Network accessline 19
terms_url: https://github.com/jwohlwend/boltz/blob/main/LICENSE - Network accessline 24
name: ColabFold MSA server (api.colabfold.com) provider: Steinegger Lab info_url: https://github.com/sokrypton/ColabFold/wiki - Network accessline 25
provider: Steinegger Lab info_url: https://github.com/sokrypton/ColabFold/wiki --- - Network accessline 26
info_url: https://github.com/sokrypton/ColabFold/wiki
skills/claude-science/borzoi/SKILL.md
- Network accessline 22
provider: Calico Life Sciences license: CC-BY-4.0 info_url: https://huggingface.co/johahi/borzoi-replicate-0 - Network accessline 23
license: CC-BY-4.0 info_url: https://huggingface.co/johahi/borzoi-replicate-0 --- - Network accessline 24
info_url: https://huggingface.co/johahi/borzoi-replicate-0
skills/claude-science/chai1/SKILL.md
- Network accessline 20
provider: Chai Discovery license: Apache-2.0 terms_url: https://github.com/chaidiscovery/chai-lab/blob/main/LICENSE - Network accessline 21
license: Apache-2.0 terms_url: https://github.com/chaidiscovery/chai-lab/blob/main/LICENSE # SKILL.md — `--use-msa-server` / `use_msa_server=True` sends the sequence - Network accessline 22
terms_url: https://github.com/chaidiscovery/chai-lab/blob/main/LICENSE - Network accessline 27
name: ColabFold MSA server (api.colabfold.com) provider: Steinegger Lab info_url: https://github.com/sokrypton/ColabFold/wiki - Network accessline 28
provider: Steinegger Lab info_url: https://github.com/sokrypton/ColabFold/wiki --- - Network accessline 29
info_url: https://github.com/sokrypton/ColabFold/wiki
skills/claude-science/compute-env-setup/references/envs_reference.md
- Access to secretsline 3
[Credential-related source omitted] - Network accessline 63
### genomics-cpu **apt:** `samtools bedtools bwa spades wget bzip2 build-essential libgomp1 libcurl4-openssl-dev libbz2-dev liblzma-dev` - Network accessline 64
**apt:** `samtools bedtools bwa spades wget bzip2 build-essential libgomp1 libcurl4-openssl-dev libbz2-dev liblzma-dev` - Network accessline 86
**apt:** git build-essential **pip_phases:** 1. `pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv` with `find_links=https://data.pyg.org/whl/torch-2.7.0+cu126.html` — **find_links not extra_index** (flat HTML, not PEP-503) - Network accessline 87
**pip_phases:** 1. `pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv` with `find_links=https://data.pyg.org/whl/torch-2.7.0+cu126.html` — **find_links not extra_index** (flat HTML, not PEP-503). Wheel URL encodes torch-min - Network accessline 88
1. `pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv` with `find_links=https://data.pyg.org/whl/torch-2.7.0+cu126.html` — **find_links not extra_index** (flat HTML, not PEP-503). Wheel URL encodes torch-minor + CUDA; this - Network accessline 96
### singlecell-gpu **base:** `pytorch/pytorch:2.7.1-cuda12.6-cudnn9-devel` — **devel** not runtime: flash-attn compiles via nvcc **apt:** git build-essential wget ninja-build - Network accessline 97
**base:** `pytorch/pytorch:2.7.1-cuda12.6-cudnn9-devel` — **devel** not runtime: flash-attn compiles via nvcc **apt:** git build-essential wget ninja-build **pip_phases:** - Network accessline 98
**apt:** git build-essential wget ninja-build - Network accessline 100
1. `scanpy anndata leidenalg igraph squidpy mudata scvi-tools>=1.2` 2. `flash-attn==2.8.0.post2` with `no_build_isolation=True` — `scgpt.tasks.embed_data` defaults `use_fast_transformer=True` 3. `git+https://github.com/bowang-lab/scGPT.git - Network accessline 101
2. `flash-attn==2.8.0.post2` with `no_build_isolation=True` — `scgpt.tasks.embed_data` defaults `use_fast_transformer=True` 3. `git+https://github.com/bowang-lab/scGPT.git gdown` with `no_deps=True` — PyPI `scgpt==0.2.4` is frozen on the le - Network accessline 102
3. `git+https://github.com/bowang-lab/scGPT.git gdown` with `no_deps=True` — PyPI `scgpt==0.2.4` is frozen on the legacy `torchtext.vocab` API; git HEAD has flash-attn-2 compat. `no_deps` because its metadata pins `scvi-tools<1.0` (a fossil - Network accessline 113
### genomics-gpu **base:** `pytorch/pytorch:2.7.1-cuda12.8-cudnn9-devel` — **devel** not runtime: flash-attn needs nvcc **apt:** git wget ca-certificates ninja-build build-essential samtools bedtools bwa libcurl4-openssl-dev libbz2-dev libl - Network accessline 114
**base:** `pytorch/pytorch:2.7.1-cuda12.8-cudnn9-devel` — **devel** not runtime: flash-attn needs nvcc **apt:** git wget ca-certificates ninja-build build-essential samtools bedtools bwa libcurl4-openssl-dev libbz2-dev liblzma-dev **pip_pha - Network accessline 115
**apt:** git wget ca-certificates ninja-build build-essential samtools bedtools bwa libcurl4-openssl-dev libbz2-dev liblzma-dev - Network accessline 118
2. `evo2 borzoi-pytorch transformers<4.51 einops intervaltree pysam biopython pandas` **env:** `CUDA_HOME=/usr/local/cuda` **run_commands:** (1) fetch mmseqs-linux-gpu binary → `/opt/mmseqs`; (2) **bake SwissProt GPU-padded DB at `/db/sprot - Network accessline 119
**env:** `CUDA_HOME=/usr/local/cuda` **run_commands:** (1) fetch mmseqs-linux-gpu binary → `/opt/mmseqs`; (2) **bake SwissProt GPU-padded DB at `/db/sprot_gpu`** — wget fasta, `mmseqs createdb` + `makepaddedseqdb`, drop unpadded. ~387MB; im - Network accessline 120
**run_commands:** (1) fetch mmseqs-linux-gpu binary → `/opt/mmseqs`; (2) **bake SwissProt GPU-padded DB at `/db/sprot_gpu`** — wget fasta, `mmseqs createdb` + `makepaddedseqdb`, drop unpadded. ~387MB; image bloat < per-job download. - Network accessline 126
### proteomics-gpu **base:** `pytorch/pytorch:2.7.1-cuda12.8-cudnn9-devel` **apt:** git wget ca-certificates build-essential - Network accessline 127
**base:** `pytorch/pytorch:2.7.1-cuda12.8-cudnn9-devel` **apt:** git wget ca-certificates build-essential **pip_phases (the chai/protenix dance — order is load-bearing):** - Network accessline 128
**apt:** git wget ca-certificates build-essential - Network accessline 151
### proteomics-jax-gpu **base:** `ghcr.io/sokrypton/colabfold:1.5.5-cuda12.2.2` — colabfold env pre-baked at `/usr/local/envs/colabfold` (py3.9) **apt:** wget ca-certificates git bzip2 - Network accessline 152
**base:** `ghcr.io/sokrypton/colabfold:1.5.5-cuda12.2.2` — colabfold env pre-baked at `/usr/local/envs/colabfold` (py3.9) **apt:** wget ca-certificates git bzip2 **pip_phases:** - Network accessline 153
**apt:** wget ca-certificates git bzip2 - Network accessline 163
### diffdock-gpu (from-source for sm_90) **base:** `pytorch/pytorch:2.4.1-cuda12.4-cudnn9-devel` (digest-pinned). Upstream `rbgcsail/diffdock` is torch 1.13.1+cu117 (no sm_90) and runs as `USER appuser` with a micromamba env. **apt:** git w - Network accessline 164
**base:** `pytorch/pytorch:2.4.1-cuda12.4-cudnn9-devel` (digest-pinned). Upstream `rbgcsail/diffdock` is torch 1.13.1+cu117 (no sm_90) and runs as `USER appuser` with a micromamba env. **apt:** git wget ca-certificates build-essential **pip - Network accessline 165
**apt:** git wget ca-certificates build-essential - Network accessline 166
**pip_phases:** 1. `torch_geometric` 2. `pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv` with `find_links=https://data.pyg.org/whl/torch-2.4.0+cu124.html` - Network accessline 167
1. `torch_geometric` 2. `pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv` with `find_links=https://data.pyg.org/whl/torch-2.4.0+cu124.html` 3. `rdkit scipy networkx biopython<2 biopandas e3nn==0.5.1 spyrmsd pyyaml pandas - Network accessline 168
2. `pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv` with `find_links=https://data.pyg.org/whl/torch-2.4.0+cu124.html` 3. `rdkit scipy networkx biopython<2 biopandas e3nn==0.5.1 spyrmsd pyyaml pandas prody fair-esm transf
skills/claude-science/customize/SKILL.md
- Instruction override signalline 208
per project**: the first delete shows a card; "Allow for this project" covers every subsequent delete in this project, including bulk teardown. Do NOT tell the user they'll see N cards for N deletes. - Instruction override signalline 210
teardown. Do NOT tell the user they'll see N cards for N deletes. - Access to secretsline 428
[Credential-related source omitted]
skills/claude-science/diffdock/SKILL.md
skills/claude-science/esmfold2/SKILL.md
- Network accessline 28
provider: Biohub license: MIT terms_url: https://github.com/Biohub/esm/blob/main/LICENSE.md - Network accessline 29
license: MIT terms_url: https://github.com/Biohub/esm/blob/main/LICENSE.md --- - Network accessline 30
terms_url: https://github.com/Biohub/esm/blob/main/LICENSE.md - Network accessline 50
uv pip install "torch>=2.5,<2.8" einops "biotite>=1.0" rdkit msgpack-numpy biopython scikit-learn brotli attrs pandas cloudpathlib httpx tenacity zstd pydssp pygtrie accelerate huggingface_hub safetensors "numpy<3" networkx sent - Network accessline 53
pygtrie accelerate huggingface_hub safetensors "numpy<3" networkx \ sentencepiece tokenizers regex packaging filelock pyyaml typing_extensions \ "transformers @ git+https://github.com/Biohub/transformers.git@3a8956fb4d4ea16b0ec8e71deef2 - Network accessline 54
sentencepiece tokenizers regex packaging filelock pyyaml typing_extensions \ "transformers @ git+https://github.com/Biohub/transformers.git@3a8956fb4d4ea16b0ec8e71deef2c2909b6a5cbf" uv pip install --no-deps "esm @ git+https://github.com/B - Network accessline 55
"transformers @ git+https://github.com/Biohub/transformers.git@3a8956fb4d4ea16b0ec8e71deef2c2909b6a5cbf" - Network accessline 56
uv pip install --no-deps "esm @ git+https://github.com/Biohub/esm.git@f652b471"
skills/claude-science/esmfold2/references/esmc.md
- Network accessline 62
Paper's atlas SAE = layer 60. Feature metadata API (alpha — endpoint may change; check biohub.ai/esm docs): `https://biohub.ai/esm/protein/api/v1alpha1/features/{idx}` returns - Network accessline 63
change; check biohub.ai/esm docs): `https://biohub.ai/esm/protein/api/v1alpha1/features/{idx}` returns `{description, top_100_uniref_ids, ...}`. 5,382 features classified as - Network accessline 64
`https://biohub.ai/esm/protein/api/v1alpha1/features/{idx}` returns
skills/claude-science/evo2/SKILL.md
- Network accessline 22
provider: Arc Institute license: Apache-2.0 terms_url: https://github.com/ArcInstitute/evo2/blob/main/LICENSE - Network accessline 23
license: Apache-2.0 terms_url: https://github.com/ArcInstitute/evo2/blob/main/LICENSE --- - Network accessline 24
terms_url: https://github.com/ArcInstitute/evo2/blob/main/LICENSE
skills/claude-science/fair-esm2/SKILL.md
- Network accessline 18
provider: Meta AI license: MIT terms_url: https://github.com/facebookresearch/esm/blob/main/LICENSE - Network accessline 19
license: MIT terms_url: https://github.com/facebookresearch/esm/blob/main/LICENSE --- - Network accessline 20
terms_url: https://github.com/facebookresearch/esm/blob/main/LICENSE
skills/claude-science/figure-composer/SKILL.md
- Dangerous shell operationline 21
exec(open("figure-composer/kernel.py").read())
skills/claude-science/figure-composer/kernel.py
skills/claude-science/figure-style/kernel.py
skills/claude-science/indication-dossier/references/00-research-standards.md
- Network accessline 7
**URL_Validation** — Only use URLs you successfully fetched or retrieved from MCP tools. NEVER construct or guess URLs. If a URL fails, note it in `anomaly_flags`. Academic journal links often break — try DOI resolver (`https://doi.org/[DOI - Network accessline 19
**PDF_Retrieval** — `WebSearch` results carry only the search-index snippet, never the PDF body. To read a PDF, call `WebFetch` on its URL: text extraction is native, so no separate download step is needed. `WebFetch` is text-only; when fig
skills/claude-science/indication-dossier/references/06-writing-style.md
skills/claude-science/learn/LICENSE.txt
skills/claude-science/ligandmpnn/SKILL.md
- Network accessline 18
name: LigandMPNN license: MIT terms_url: https://github.com/dauparas/LigandMPNN/blob/main/LICENSE - Network accessline 19
license: MIT terms_url: https://github.com/dauparas/LigandMPNN/blob/main/LICENSE --- - Network accessline 20
terms_url: https://github.com/dauparas/LigandMPNN/blob/main/LICENSE - Network accessline 41
pip install torch numpy biopython ProDy ml_collections dm-tree git clone --depth 1 https://github.com/dauparas/LigandMPNN.git ligandmpnn - Network accessline 42
git clone --depth 1 https://github.com/dauparas/LigandMPNN.git ligandmpnn
skills/claude-science/literature-review/SKILL.md
- Access to secretsline 7
[Credential-related source omitted] - Network accessline 15
- kind: service name: Crossref info_url: https://www.crossref.org/documentation/retrieve-metadata/ - Network accessline 16
name: Crossref info_url: https://www.crossref.org/documentation/retrieve-metadata/ privacy_url: https://www.crossref.org/operations-and-sustainability/privacy/ - Network accessline 17
info_url: https://www.crossref.org/documentation/retrieve-metadata/ - Network accessline 18
privacy_url: https://www.crossref.org/operations-and-sustainability/privacy/ - Network accessline 19
- kind: service name: OpenAlex terms_url: https://openalex.org/OpenAlex_termsofservice.pdf - Network accessline 20
name: OpenAlex terms_url: https://openalex.org/OpenAlex_termsofservice.pdf privacy_url: https://openalex.org/OpenAlex_privacy_policy.pdf - Network accessline 21
terms_url: https://openalex.org/OpenAlex_termsofservice.pdf - Network accessline 22
privacy_url: https://openalex.org/OpenAlex_privacy_policy.pdf - Dangerous shell operationline 32
exec(open("<this skill's directory>/kernel.py").read()) - Network accessline 36
Configuration is via environment variables, not a host — no LLM key is involved: - `OPENALEX_API_KEY` — required for the OpenAlex-backed steps (`search_openalex`, `expand_citations`); free at https://openalex.org/settings/api. - Network accessline 37
[Credential-related source omitted] - Access to secretsline 37
[Credential-related source omitted] - Network accessline 78
Cite inline as a markdown link — `[Author Year](https://doi.org/10.xxxx/xxxxxx)` — so the rendered prose reads `(Author Year)` and the DOI rides in the href where a reader can click it and a regex can still extract it. If the DOI itself con
skills/claude-science/literature-review/kernel.py
- Executable scriptline 1
Executable file - Dangerous shell operationline 11
exec(open("<this-skill-dir>/kernel.py").read()) - Network accessline 14
[Credential-related source omitted] - Access to secretsline 14
[Credential-related source omitted] - Network accessline 15
/ expand_citations); free at https://openalex.org/settings/api HOST_USER_EMAIL optional contact email for the CrossRef/doi.org polite - Network accessline 16
https://openalex.org/settings/api - Access to secretsline 43
[Credential-related source omitted] - Access to secretsline 47
[Credential-related source omitted] - Access to secretsline 65
[Credential-related source omitted] - Access to secretsline 68
[Credential-related source omitted] - Network accessline 72
"OpenAlex requires a free API key and none is set. Export " "OPENALEX_API_KEY and re-run — keyless calls are unsupported. Get a " "free key at https://openalex.org/settings/api, or skip the " - Network accessline 73
[Credential-related source omitted] - Access to secretsline 73
[Credential-related source omitted] - Network accessline 74
"free key at https://openalex.org/settings/api, or skip the " - Access to secretsline 81
[Credential-related source omitted] - Network accessline 88
[Credential-related source omitted] - Access to secretsline 88
[Credential-related source omitted] - Network accessline 89
+ urllib.parse.quote(key, safe="")) req = urllib.request.Request( url, headers={"User-Agent": "ClaudeScience-literature-review/1.0"} - Network accessline 90
req = urllib.request.Request( - Network accessline 92
) try: with urllib.request.urlopen(req, timeout=timeout): - Network accessline 93
try: with urllib.request.urlopen(req, timeout=timeout): return True - Network accessline 94
with urllib.request.urlopen(req, timeout=timeout): - Access to secretsline 105
[Credential-related source omitted] - Network accessline 119
[Credential-related source omitted] - Access to secretsline 119
[Credential-related source omitted] - Network accessline 120
for attempt in (0, 1): req = urllib.request.Request( full, headers={"User-Agent": "ClaudeScience-literature-review/1.0"} - Network accessline 121
req = urllib.request.Request( - Network accessline 123
) try: with urllib.request.urlopen(req, timeout=timeout) as r: - Network accessline 124
try: with urllib.request.urlopen(req, timeout=timeout) as r: return json.loads(r.read().decode("utf-8")) - Network accessline 125
with urllib.request.urlopen(req, timeout=timeout) as r: - Network accessline 130
"OpenAlex rejected the API key (HTTP %d). Re-check it " "under Customize → Credentials → OpenAlex against " "https://openalex.org/settings/api — do not retry " - Network accessline 131
"under Customize → Credentials → OpenAlex against " "https://openalex.org/settings/api — do not retry " "anonymously." % e.code - Network accessline 132
"https://openalex.org/settings/api — do not retry " - Network accessline 143
"Re-check it under Customize → Credentials → " "OpenAlex against " "https://openalex.org/settings/api — do not retry " - Network accessline 144
"OpenAlex against " "https://openalex.org/settings/api — do not retry " "anonymously." - Network accessline 145
"https://openalex.org/settings/api — do not retry " - Network accessline 172
ua = ua.encode("ascii", "ignore").decode("ascii") for attempt in (0, 1): req = urllib.request.Request(url, headers={"User-Agent": ua}) - Network accessline 173
for attempt in (0, 1): req = urllib.request.Request(url, headers={"User-Agent": ua}) try: - Network accessline 174
req = urllib.request.Request(url, headers={"User-Agent": ua}) - Network accessline 175
try: with urllib.request.urlopen(req, timeout=timeout) as r: return json.loads(r.read().decode("utf-8")) - Network accessline 176
with urllib.request.urlopen(req, timeout=timeout) as r: - Network accessline 247
opener = urllib.request.build_opener(NoRedirect) for attempt in (0, 1): req = urllib.request.Request(url, headers={"User-Agent": ua}, method="HEAD") - Network accessline 248
for attempt in (0, 1): req = urllib.request.Request(url, headers={"User-Agent": ua}, method="HEAD") try: - Network accessline 249
req = urllib.request.Request(url, headers={"User-Agent": ua}, method="HEAD") - Network accessline 284
continue enc = quote_doi_path(d) j = litrev_get(f"https://api.crossref.org/works/{enc}") - Network accessline 285
enc = quote_doi_path(d) j = litrev_get(f"https://api.crossref.org/works/{enc}") time.sleep(0.06) - Network accessline 286
j = litrev_get(f"https://api.crossref.org/works/{enc}") - Network accessline 309
code = litrev_head(f"https://doi.org/{enc}") - Network accessline 323
q = urllib.parse.quote(ref_string) j = litrev_get(f"https://api.crossref.org/works?query.bibliographic={q}&rows=1") - Network accessline 324
j = litrev_get(f"https://api.crossref.org/works?query.bibliographic={q}&rows=1") - Network accessline 346
flt = f"&filter={filters}" if filters else "" j = litrev_openalex_get( f"https://api.openalex.org/works?search={q}&per-page={min(n, 25)}" - Network accessline 347
j = litrev_openalex_get( f"https://api.openalex.org/works?search={q}&per-page={min(n, 25)}" f"&sort=cited_by_count:desc{flt}" - Network accessline 348
f"https://api.openalex.org/works?search={q}&per-page={min(n, 25)}" - Network accessline 355
out.append( { "doi": (w.get("doi") or "").replace("https://doi.org/", ""), - Network accessline 356
{ "doi": (w.get("doi") or "").replace("https://doi.org/", ""), "title": w.get("title"), - Network accessline 357
"doi": (w.get("doi") or "").replace("https://doi.org/", ""), - Network accessline 381
enc = quote_doi_path(doi) work = litrev_openalex_get( f"https://api.openalex.org/works/doi:{enc}?select=id" - Network accessline 382
work = litrev_openalex_get( f"https://api.openalex.org/works/doi:{enc}?select=id" ) - Network accessline 383
f"https://api.openalex.org/works/doi:{enc}?select=id" - Network accessline 392
out.append( { "doi": (w.get("doi") or "").replace("https://doi.org/", ""), - Network accessline 393
{ "doi": (w.get("doi") or "").replace("https://doi.org/", ""), "title": w.get("title"), - Network accessline 394
"doi": (w.get("doi") or "").replace("https://doi.org/", ""), - Network accessline 403
def _list(filter_expr: str, n: int) -> list[dict]: base = ( f"https://api.openalex.org/works?filter={filter_expr}" - Network accessline 404
base = ( f"https://api.openalex.org/works?filter={filter_expr}" f"&sort=cited_by_count:desc&per-page={min(n, 100)}" - Network accessline 405
f"https://api.openalex.org/works?filter={filter_expr}" - Network accessline 465
if re.search(r"(DOIs?\s+(were\s+)?verif|verified against (CrossRef|PubMed)|no retraction|current as of)", draft, re.I): issues.append({"code": "PROCNOTE", "note": "process-narration line present; delete it"}) if re.search(r"\]\( - Network accessline 466
issues.append({"code": "PROCNOTE", "note": "process-narration line present; delete it"}) if re.search(r"\]\(https://doi\.org/[^)\s]*\([^)\s]*\)", draft): issues.append({"code": "PARENDOI", "note": "DOI href contains literal ( ); - Network accessline 467
if re.search(r"\]\(https://doi\.org/[^)\s]*\([^)\s]*\)", draft):
skills/claude-science/managed-model-endpoints/SKILL.md
- Network accessline 33
compute_provider(provider="boltz2-service", code=""" import requests r = requests.post(BASE_URL + "/v1/infer", json=payload) - Network accessline 34
import requests r = requests.post(BASE_URL + "/v1/infer", json=payload) """) - Network accessline 35
r = requests.post(BASE_URL + "/v1/infer", json=payload) - Access to secretsline 42
[Credential-related source omitted] - Access to secretsline 63
[Credential-related source omitted] - Access to secretsline 66
[Credential-related source omitted] - Network accessline 75
name="boltz2-service", # <model>-service -- descriptive, never the # bare model name (collides/ambiguous) url=f"http://127.0.0.1:{port}", # LITERAL 127.0.0.1 -- `localhost` rejected - Network accessline 76
# bare model name (collides/ambiguous) url=f"http://127.0.0.1:{port}", # LITERAL 127.0.0.1 -- `localhost` rejected credential="NVIDIA_API_KEY", # the family credential NAME, never a value - Network accessline 77
url=f"http://127.0.0.1:{port}", # LITERAL 127.0.0.1 -- `localhost` rejected - Access to secretsline 78
[Credential-related source omitted] - Access to secretsline 107
[Credential-related source omitted] - Access to secretsline 130
[Credential-related source omitted] - Access to secretsline 133
[Credential-related source omitted] - Dangerous shell operationline 139
docker rm -f boltz2-service # stale port mapping -- recreate below - Access to secretsline 144
[Credential-related source omitted] - Dangerous shell operationline 164
docker rm -f boltz2-service - Access to secretsline 181
[Credential-related source omitted]
skills/claude-science/openfold3/SKILL.md
- Network accessline 18
provider: OpenFold Consortium license: Apache-2.0 terms_url: https://github.com/aqlaboratory/openfold-3/blob/main/LICENSE - Network accessline 19
license: Apache-2.0 terms_url: https://github.com/aqlaboratory/openfold-3/blob/main/LICENSE # SKILL.md — `--use-msa-server` DEFAULTS to true (MSA server is - Network accessline 20
terms_url: https://github.com/aqlaboratory/openfold-3/blob/main/LICENSE - Network accessline 25
name: ColabFold MSA server (api.colabfold.com) provider: Steinegger Lab info_url: https://github.com/sokrypton/ColabFold/wiki - Network accessline 26
provider: Steinegger Lab info_url: https://github.com/sokrypton/ColabFold/wiki --- - Network accessline 27
info_url: https://github.com/sokrypton/ColabFold/wiki
skills/claude-science/paper-narrative/SKILL.md
- Dangerous shell operationline 20
exec(open("paper-narrative/kernel.py").read()) # path to this skill's kernel.py
skills/claude-science/paper-narrative/kernel.py
skills/claude-science/pdf-explore/SKILL.md
- Dangerous shell operationline 27
exec(open("skills/claude-science/pdf-explore/kernel.py").read())
skills/claude-science/pdf-explore/kernel.py
skills/claude-science/product-self-knowledge/SKILL.md
- Network accessline 24
- **Claude API & General:** https://docs.claude.com/en/docs_site_map.md - Network accessline 25
- **Claude Code:** https://docs.claude.com/en/docs/claude-code/claude_code_docs_map.md - Network accessline 31
- **Claude.ai Help Center:** https://support.claude.com - Network accessline 49
- Documentation: https://docs.claude.com/en/api/overview - Network accessline 50
- Docs Map: https://docs.claude.com/en/docs_site_map.md - Network accessline 54
- Documentation: https://docs.claude.com/en/docs/claude-code/overview - Network accessline 55
- Docs Map: https://docs.claude.com/en/docs/claude-code/claude_code_docs_map.md - Network accessline 56
- npm Package: https://www.npmjs.com/package/@anthropic-ai/claude-code - Network accessline 60
- Support Center: https://support.claude.com - Network accessline 64
- Product News: https://www.anthropic.com/news - Network accessline 65
- Enterprise Sales: https://www.anthropic.com/contact-sales
skills/claude-science/proteinmpnn/SKILL.md
- Network accessline 17
name: ProteinMPNN license: MIT terms_url: https://github.com/dauparas/ProteinMPNN/blob/main/LICENSE - Network accessline 18
license: MIT terms_url: https://github.com/dauparas/ProteinMPNN/blob/main/LICENSE --- - Network accessline 19
terms_url: https://github.com/dauparas/ProteinMPNN/blob/main/LICENSE - Network accessline 40
pip install torch numpy # if not already present git clone --depth 1 https://github.com/dauparas/ProteinMPNN.git proteinmpnn - Network accessline 41
git clone --depth 1 https://github.com/dauparas/ProteinMPNN.git proteinmpnn
skills/claude-science/remote-compute-modal/SKILL.md
- Network accessline 734
| what the output says | under an allowlist it means | under No network it means | |---|---|---| | `Connection reset by peer` / `ECONNRESET` on an `https://` URL | the host is not on the sandbox's list | (everything already fails earlier, a - Network accessline 735
|---|---|---| | `Connection reset by peer` / `ECONNRESET` on an `https://` URL | the host is not on the sandbox's list | (everything already fails earlier, at DNS) | | `Network is unreachable` / `ENETUNREACH` | the request was not TLS on po - Network accessline 736
| `Connection reset by peer` / `ECONNRESET` on an `https://` URL | the host is not on the sandbox's list | (everything already fails earlier, at DNS) | | `Network is unreachable` / `ENETUNREACH` | the request was not TLS on port 443 — plain - Network accessline 737
| `Network is unreachable` / `ENETUNREACH` | the request was not TLS on port 443 — plain `http://`, a non-443 port, or a raw socket — and the filter refuses those even for an *allowed* domain | no route to anything | - Network accessline 747
(or, for an env's documented commands, declare it in `META["egress_domains"]`). A non-TLS / non-443 failure (`apt-get`, an `http://` mirror, a raw socket): no allowlist edit can fix it — use - Network accessline 748
`META["egress_domains"]`). A non-TLS / non-443 failure (`apt-get`, an `http://` mirror, a raw socket): no allowlist edit can fix it — use HTTPS or do that install at image-build time, which is not fenced. - Network accessline 749
`http://` mirror, a raw socket): no allowlist edit can fix it — use - Network accessline 761
one domain is missing trades their whole egress posture for a one-line allowlist edit — name the domain instead. And probing from inside the job ("let me curl a few sites to test the network") only reproduces the - Network accessline 762
allowlist edit — name the domain instead. And probing from inside the job ("let me curl a few sites to test the network") only reproduces the failure you already have; the `[egress]` line and `job.egress` already - Network accessline 763
job ("let me curl a few sites to test the network") only reproduces the - Network accessline 770
only** — once an allowlist is set, plain-HTTP and non-443 traffic from the sandbox is refused regardless of the domain (we verified this live: `http://example.com` is refused even with `example.com` on the list). A - Network accessline 771
the sandbox is refused regardless of the domain (we verified this live: `http://example.com` is refused even with `example.com` on the list). A tool that fetches over `http://` will fail under any allowlist even if - Network accessline 772
`http://example.com` is refused even with `example.com` on the list). A - Network accessline 773
tool that fetches over `http://` will fail under any allowlist even if
skills/claude-science/remote-compute-modal/env-setup.md
- Dangerous shell operationline 44
`modal` and `app`. It exec()s the named file, calls `build()`, builds the - Network accessline 224
"dm-tree") .run_commands( "git clone https://github.com/dauparas/LigandMPNN.git /app/ligandmpnn" - Network accessline 225
.run_commands( "git clone https://github.com/dauparas/LigandMPNN.git /app/ligandmpnn" " && git -C /app/ligandmpnn checkout 26ec57ac976ade5379920dbd43c7f97a91cf82de", # pinned 2026-06-28 - Network accessline 226
"git clone https://github.com/dauparas/LigandMPNN.git /app/ligandmpnn" - Network accessline 267
vol = modal.Volume.from_name("af2-params", create_if_missing=True) sb = modal.Sandbox.create("bash", "-lc", "wget -q -O /params/af2.tar https://storage.googleapis.com/alphafold/" - Network accessline 268
sb = modal.Sandbox.create("bash", "-lc", "wget -q -O /params/af2.tar https://storage.googleapis.com/alphafold/" "alphafold_params_2022-12-06.tar" - Network accessline 269
"wget -q -O /params/af2.tar https://storage.googleapis.com/alphafold/" - Dangerous shell operationline 273
" && tar xf /params/af2.tar -C /params && rm /params/af2.tar",
skills/claude-science/remote-compute-modal/envs/chemistry_gpu.py
- Executable scriptline 1
Executable file
skills/claude-science/remote-compute-modal/envs/esmfold2_gpu.py
- Executable scriptline 1
Executable file - Network accessline 32
_TRANSFORMERS_PIN = ( "transformers @ git+https://github.com/Biohub/transformers.git" - Network accessline 33
"transformers @ git+https://github.com/Biohub/transformers.git" - Network accessline 34
"@3a8956fb4d4ea16b0ec8e71deef2c2909b6a5cbf" ) _ESM_PIN = "esm @ git+https://github.com/Biohub/esm.git@f652b471d29da828b31e9b7a9cf7d0a7803240f5" - Network accessline 36
_ESM_PIN = "esm @ git+https://github.com/Biohub/esm.git@f652b471d29da828b31e9b7a9cf7d0a7803240f5"
skills/claude-science/remote-compute-modal/envs/genomics_evo2_gpu.py
- Executable scriptline 1
Executable file
skills/claude-science/remote-compute-modal/envs/proteomics_boltz_gpu.py
- Executable scriptline 1
Executable file
skills/claude-science/remote-compute-modal/envs/proteomics_gpu.py
- Executable scriptline 1
Executable file - Network accessline 56
) .run_commands( "wget -q https://github.com/soedinglab/MMseqs2/releases/download/" - Network accessline 57
.run_commands( "wget -q https://github.com/soedinglab/MMseqs2/releases/download/" "18-8cc5c/mmseqs-linux-gpu.tar.gz -O /tmp/mm.tar.gz", - Network accessline 58
"wget -q https://github.com/soedinglab/MMseqs2/releases/download/" - Network accessline 65
"git clone https://github.com/dauparas/ProteinMPNN.git"
skills/claude-science/remote-compute-modal/envs/proteomics_jax_gpu.py
- Executable scriptline 1
Executable file
skills/claude-science/remote-compute-modal/envs/proteomics_openfold_gpu.py
- Executable scriptline 1
Executable file - Network accessline 37
} _CKPT_URL = ( "https://openfold3-data.s3.amazonaws.com/openfold3-parameters/of3-p2-155k.pt" - Network accessline 38
_CKPT_URL = ( "https://openfold3-data.s3.amazonaws.com/openfold3-parameters/of3-p2-155k.pt" ) - Network accessline 39
"https://openfold3-data.s3.amazonaws.com/openfold3-parameters/of3-p2-155k.pt" - Dangerous shell operationline 117
"{ rm -f of3-p2-155k.pt; "
skills/claude-science/remote-compute-modal/envs/proteomics_rfd_diffdock_gpu.py
- Executable scriptline 1
Executable file - Network accessline 47
_DIFFDOCK_SHA = "9a22cbcbc7612c7565c80e8399d9be298971f156" # v1.1.3 _RFD_SHA = "2d0c003df46b9db41d119321f15403dec3716cd9" _WEIGHTS_URL = "https://files.ipd.uw.edu/pub/RFdiffusion/" - Network accessline 49
_WEIGHTS_URL = "https://files.ipd.uw.edu/pub/RFdiffusion/" - Network accessline 70
"torch-sparse==0.6.18", "torch-cluster==1.6.3", find_links="https://data.pyg.org/whl/torch-2.1.0+cu121.html", - Network accessline 71
"torch-cluster==1.6.3", find_links="https://data.pyg.org/whl/torch-2.1.0+cu121.html", ) - Network accessline 72
find_links="https://data.pyg.org/whl/torch-2.1.0+cu121.html", - Network accessline 74
.pip_install( "dgl==1.1.3", find_links="https://data.dgl.ai/wheels/cu121/repo.html", - Network accessline 75
"dgl==1.1.3", find_links="https://data.dgl.ai/wheels/cu121/repo.html", ) - Network accessline 76
find_links="https://data.dgl.ai/wheels/cu121/repo.html", - Network accessline 90
.run_commands( "git init /opt/diffdock && cd /opt/diffdock && " "git remote add origin https://github.com/gcorso/DiffDock.git && " - Network accessline 91
"git init /opt/diffdock && cd /opt/diffdock && " "git remote add origin https://github.com/gcorso/DiffDock.git && " f"git fetch --depth 1 origin {_DIFFDOCK_SHA} && " - Network accessline 92
"git remote add origin https://github.com/gcorso/DiffDock.git && " f"git fetch --depth 1 origin {_DIFFDOCK_SHA} && " "git checkout FETCH_HEAD", - Network accessline 95
"git init /opt/rfd && cd /opt/rfd && " "git remote add origin " "https://github.com/RosettaCommons/RFdiffusion.git && " - Network accessline 96
"git remote add origin " "https://github.com/RosettaCommons/RFdiffusion.git && " f"git fetch --depth 1 origin {_RFD_SHA} && " - Network accessline 97
"https://github.com/RosettaCommons/RFdiffusion.git && " f"git fetch --depth 1 origin {_RFD_SHA} && " "git checkout FETCH_HEAD", - Network accessline 112
# RFdiffusion model weights from files.ipd.uw.edu. # `wget -qc` resumes partial downloads (so a prior interrupted fetch is - Network accessline 113
# `wget -qc` resumes partial downloads (so a prior interrupted fetch is - Network accessline 127
' python -c "import zipfile,sys; ' " sys.exit(0 if zipfile.is_zipfile('$f') else 1)\" 2>/dev/null || " f' {{ rm -f "$f"; wget -qc {_WEIGHTS_URL}"$f"; }}; done', - Network accessline 128
" sys.exit(0 if zipfile.is_zipfile('$f') else 1)\" 2>/dev/null || " f' {{ rm -f "$f"; wget -qc {_WEIGHTS_URL}"$f"; }}; done', ) - Dangerous shell operationline 129
f' {{ rm -f "$f"; wget -qc {_WEIGHTS_URL}"$f"; }}; done', - Network accessline 129
f' {{ rm -f "$f"; wget -qc {_WEIGHTS_URL}"$f"; }}; done',
skills/claude-science/remote-compute-modal/envs/singlecell_gpu.py
skills/claude-science/remote-compute-modal/provider.py
- Executable scriptline 1
Executable file - Access to secretsline 53
[Credential-related source omitted] - Dangerous shell operationline 86
the warm-reuse ``rm -rf /work && mkdir && tar`` window where - Dangerous shell operationline 90
job whose cleanup deleted ``run.sh`` (e.g. ``rm -rf ./*`` in /work) - Access to secretsline 146
[Credential-related source omitted] - Access to secretsline 147
[Credential-related source omitted] - Access to secretsline 200
[Credential-related source omitted] - Access to secretsline 202
[Credential-related source omitted] - Access to secretsline 204
[Credential-related source omitted] - Dangerous shell operationline 332
def exec(self, sandbox_id: str, argv: list[str], *, stdin: Iterable[bytes] | None = None, - Access to secretsline 590
[Credential-related source omitted] - Access to secretsline 592
[Credential-related source omitted] - Network accessline 709
"A concurrent limit frees up when idle handles and sandboxes are " "closed — they count against it. A spend limit doesn't: accrued " "spend stays accrued — raise it at https://modal.com/settings/plan, " - Network accessline 710
"closed — they count against it. A spend limit doesn't: accrued " "spend stays accrued — raise it at https://modal.com/settings/plan, " f"or lower gpu/cpu. ({raw})", - Network accessline 711
"spend stays accrued — raise it at https://modal.com/settings/plan, " - Network accessline 723
"transient", "Modal API was unreachable or timed out — usually transient. Retry in " f"~30s; if it persists check https://status.modal.com. ({raw})", - Network accessline 724
"Modal API was unreachable or timed out — usually transient. Retry in " f"~30s; if it persists check https://status.modal.com. ({raw})", ) - Network accessline 725
f"~30s; if it persists check https://status.modal.com. ({raw})",
skills/claude-science/remote-compute-ssh/SKILL.md
- Access to secretsline 81
[Credential-related source omitted]
skills/claude-science/scgpt/SKILL.md
- Network accessline 24
name: scGPT provider: Wang Lab (University of Toronto) info_url: https://github.com/bowang-lab/scGPT - Network accessline 25
provider: Wang Lab (University of Toronto) info_url: https://github.com/bowang-lab/scGPT --- - Network accessline 26
info_url: https://github.com/bowang-lab/scGPT
skills/claude-science/scvi-tools/SKILL.md
- Dangerous shell operationline 35
exec(open("scvi-tools/kernel.py").read()) # path to this skill's kernel.py
skills/claude-science/scvi-tools/kernel.py
- Executable scriptline 1
Executable file
skills/claude-science/skill-creator/assets/eval_review.html
- Network accessline 5
<meta name="viewport" content="width=device-width, initial-scale=1.0" /> <title>Eval Set Review - __SKILL_NAME_PLACEHOLDER__</title> <link rel="preconnect" href="https://fonts.googleapis.com" /> - Network accessline 6
<title>Eval Set Review - __SKILL_NAME_PLACEHOLDER__</title> <link rel="preconnect" href="https://fonts.googleapis.com" /> <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin /> - Network accessline 7
<link rel="preconnect" href="https://fonts.googleapis.com" /> - Network accessline 8
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin /> - Network accessline 9
<link href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap" rel="stylesheet" - Network accessline 10
href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap"
skills/claude-science/skill-creator/eval-viewer/generate_review.py
skills/claude-science/skill-creator/eval-viewer/viewer.html
- Network accessline 5
<meta name="viewport" content="width=device-width, initial-scale=1.0" /> <title>Eval Review</title> <link rel="preconnect" href="https://fonts.googleapis.com" /> - Network accessline 6
<title>Eval Review</title> <link rel="preconnect" href="https://fonts.googleapis.com" /> <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin /> - Network accessline 7
<link rel="preconnect" href="https://fonts.googleapis.com" /> - Network accessline 8
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin /> - Network accessline 9
<link href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap" rel="stylesheet" - Network accessline 10
href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap" - Network accessline 12
/> <script src="https://cdn.sheetjs.com/xlsx-0.20.3/package/dist/xlsx.full.min.js" - Network accessline 13
<script src="https://cdn.sheetjs.com/xlsx-0.20.3/package/dist/xlsx.full.min.js" integrity="sha384-EnyY0/GSHQGSxSgMwaIPzSESbqoOLSexfnSMN2AP+39Ckmn92stwABZynq1JyzdT" - Network accessline 14
src="https://cdn.sheetjs.com/xlsx-0.20.3/package/dist/xlsx.full.min.js" - Network accessline 757
if (!hasPrevious) { try { const resp = await fetch("/api/feedback"); - Network accessline 758
try { const resp = await fetch("/api/feedback"); const data = await resp.json(); - Network accessline 759
const resp = await fetch("/api/feedback"); - Network accessline 1144
fetch("/api/feedback", { - Network accessline 1185
2, ); fetch("/api/feedback", { - Network accessline 1186
); fetch("/api/feedback", { method: "POST", - Network accessline 1187
fetch("/api/feedback", {
skills/claude-science/skill-creator/scripts/__init__.py
- Executable scriptline 1
Executable file
skills/claude-science/skill-creator/scripts/aggregate_benchmark.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/claude-science/skill-creator/scripts/generate_report.py
- Executable scriptline 1
#!/usr/bin/env python3 - Network accessline 37
<meta charset="utf-8"> """ + refresh_tag + """ <title>""" + title_prefix + """Skill Description Optimization</title> <link rel="preconnect" href="https://fonts.googleapis.com"> - Network accessline 38
""" + refresh_tag + """ <title>""" + title_prefix + """Skill Description Optimization</title> <link rel="preconnect" href="https://fonts.googleapis.com"> <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin> - Network accessline 39
<link rel="preconnect" href="https://fonts.googleapis.com"> - Network accessline 40
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin> - Network accessline 41
<link href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap" rel="stylesheet">
skills/claude-science/skill-creator/scripts/improve_description.py
skills/claude-science/skill-creator/scripts/package_skill.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/claude-science/skill-creator/scripts/quick_validate.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/claude-science/skill-creator/scripts/run_eval.py
skills/claude-science/skill-creator/scripts/run_loop.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/claude-science/skill-creator/scripts/utils.py
- Executable scriptline 1
Executable file
skills/claude-science/solublempnn/SKILL.md
- Network accessline 19
name: SolubleMPNN license: MIT terms_url: https://github.com/dauparas/ProteinMPNN/blob/main/LICENSE - Network accessline 20
license: MIT terms_url: https://github.com/dauparas/ProteinMPNN/blob/main/LICENSE --- - Network accessline 21
terms_url: https://github.com/dauparas/ProteinMPNN/blob/main/LICENSE - Network accessline 42
pip install torch numpy # if not already present git clone --depth 1 https://github.com/dauparas/ProteinMPNN.git proteinmpnn - Network accessline 43
git clone --depth 1 https://github.com/dauparas/ProteinMPNN.git proteinmpnn
skills/claude-science/using-model-endpoint/SKILL.md
skills/claude-science/using-model-endpoint/provider.py
- Executable scriptline 1
Executable file - Access to secretsline 57
[Credential-related source omitted] - Access to secretsline 59
[Credential-related source omitted] - Access to secretsline 61
[Credential-related source omitted] - Access to secretsline 64
[Credential-related source omitted] - Access to secretsline 67
[Credential-related source omitted] - Access to secretsline 74
[Credential-related source omitted] - Dangerous shell operationline 99
def exec(
skills/claude-science/web-artifacts-builder/LICENSE.txt
skills/claude-science/web-artifacts-builder/SKILL.md
- Network accessline 74
- **shadcn/ui components**: https://ui.shadcn.com/docs/components
skills/claude-science/web-artifacts-builder/scripts/bundle-artifact.sh
skills/claude-science/web-artifacts-builder/scripts/init-artifact.sh
skills/scientific-visualization/assets/color_palettes.py
skills/scientific-visualization/references/color_palettes.md
- Network accessline 233
### Online Simulators - **Coblis**: https://www.color-blindness.com/coblis-color-blindness-simulator/ - Network accessline 234
- **Coblis**: https://www.color-blindness.com/coblis-color-blindness-simulator/ - Network accessline 344
- **ColorBrewer**: http://colorbrewer2.org/ - Choose palettes by colorblind-safe option - Network accessline 345
- **Paul Tol's palettes**: https://personal.sron.nl/~pault/ - Network accessline 346
- **Okabe-Ito palette origin**: "Color Universal Design" (Okabe & Ito, 2008) - **Matplotlib colormaps**: https://matplotlib.org/stable/tutorials/colors/colormaps.html - **Seaborn palettes**: https://seaborn.pydata.org/tutorial/color_palette - Network accessline 347
- **Matplotlib colormaps**: https://matplotlib.org/stable/tutorials/colors/colormaps.html - Network accessline 348
- **Seaborn palettes**: https://seaborn.pydata.org/tutorial/color_palettes.html
skills/scientific-visualization/references/matplotlib_examples.md
skills/scientific-visualization/scripts/figure_export.py
- Executable scriptline 1
#!/usr/bin/env python3
skills/scientific-visualization/scripts/style_presets.py
- Executable scriptline 1
#!/usr/bin/env python3
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Declared and executed: 5 facts
- Contacts openfold3-data.s3.amazonaws.com, which SKILL.md, the README and the plugin manifest do not mentionremote-compute-modal/envs/proteomics_openfold_gpu.py:39High confidence
- Contacts files.ipd.uw.edu, which SKILL.md, the README and the plugin manifest do not mentionremote-compute-modal/envs/proteomics_rfd_diffdock_gpu.py:49High confidence
- Contacts data.pyg.org, which SKILL.md, the README and the plugin manifest do not mentionremote-compute-modal/envs/proteomics_rfd_diffdock_gpu.py:72High confidence
- Contacts modal.com, which SKILL.md, the README and the plugin manifest do not mentionremote-compute-modal/provider.py:711High confidence
- Contacts fonts.gstatic.com, which SKILL.md, the README and the plugin manifest do not mentionskill-creator/scripts/generate_report.py:40High confidence
Signal confidence: 373 facts
- Network access in prose of documentationalgorithmic-art/LICENSE.txt:2Low confidence
- Network access in prose of documentationalgorithmic-art/LICENSE.txt:4Low confidence
- Network access in prose of documentationalgorithmic-art/LICENSE.txt:196Low confidence
- Network access in code of the SKILL.mdalgorithmic-art/SKILL.md:278High confidence
- Network access in code of the SKILL.mdalgorithmic-art/SKILL.md:279High confidence
- Network access in code of the SKILL.mdalgorithmic-art/SKILL.md:280High confidence
- Executable script in a comment of a file that runsalgorithmic-art/templates/generator_template.js:1Low confidence
- Network access in the frontmatter of the SKILL.mdalphafold2/SKILL.md:21Low confidence
- Network access in the frontmatter of the SKILL.mdalphafold2/SKILL.md:22Low confidence
- Network access in the frontmatter of the SKILL.mdalphafold2/SKILL.md:23Low confidence
- Network access in the frontmatter of the SKILL.mdalphafold2/SKILL.md:28Low confidence
- Network access in the frontmatter of the SKILL.mdalphafold2/SKILL.md:29Low confidence
- Network access in the frontmatter of the SKILL.mdalphafold2/SKILL.md:30Low confidence
- Network access in the frontmatter of the SKILL.mdboltz/SKILL.md:17Low confidence
- Network access in the frontmatter of the SKILL.mdboltz/SKILL.md:18Low confidence
- Network access in the frontmatter of the SKILL.mdboltz/SKILL.md:19Low confidence
- Network access in the frontmatter of the SKILL.mdboltz/SKILL.md:24Low confidence
- Network access in the frontmatter of the SKILL.mdboltz/SKILL.md:25Low confidence
- Network access in the frontmatter of the SKILL.mdboltz/SKILL.md:26Low confidence
- Network access in the frontmatter of the SKILL.mdborzoi/SKILL.md:22Low confidence
- Network access in the frontmatter of the SKILL.mdborzoi/SKILL.md:23Low confidence
- Network access in the frontmatter of the SKILL.mdborzoi/SKILL.md:24Low confidence
- Network access in the frontmatter of the SKILL.mdchai1/SKILL.md:20Low confidence
- Network access in the frontmatter of the SKILL.mdchai1/SKILL.md:21Low confidence
- Network access in the frontmatter of the SKILL.mdchai1/SKILL.md:22Low confidence
- Network access in the frontmatter of the SKILL.mdchai1/SKILL.md:27Low confidence
- Network access in the frontmatter of the SKILL.mdchai1/SKILL.md:28Low confidence
- Network access in the frontmatter of the SKILL.mdchai1/SKILL.md:29Low confidence
- Access to secrets in prose of documentation: a variable, placeholder or call, not a stored valuecompute-env-setup/references/envs_reference.md:3Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:63Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:64Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:86Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:87Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:88Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:96Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:97Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:98Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:100Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:101Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:102Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:113Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:114Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:115Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:118Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:119Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:120Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:126Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:127Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:128Low confidence
- Network access in prose of documentationcompute-env-setup/references/envs_reference.md:151Low confidence
and 323 more
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Compatible | Standard SKILL.md skills/claude-science/algorithmic-art/SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md skills/claude-science/algorithmic-art/SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md skills/claude-science/algorithmic-art/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md skills/claude-science/algorithmic-art/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md skills/claude-science/algorithmic-art/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md skills/claude-science/algorithmic-art/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
Explore alternatives and comparisons
Add this badge to your README
It updates every night and links to this page.
[](https://hotskills.dev/skill/hughyau-academicforge/)