### Install Dependencies and Run Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/timesfm-forecasting/examples/global-temperature/README.md

Installs necessary Python packages using uv pip and executes the forecasting example script. Ensure you are in the correct directory before running.

```bash
# Install dependencies
uv pip install "timesfm[torch]" matplotlib pandas numpy

# Run the complete example
cd skills/timesfm-forecasting/examples/global-temperature
./run_example.sh
```

--------------------------------

### Start the Application

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/markdown-mermaid-writing/templates/project_documentation.md

Run this command to start the application after installing dependencies and configuring the environment.

```bash
[package-manager] run dev
```

--------------------------------

### Install and Setup Latch SDK

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/latchbio-integration/SKILL.md

Installs the Latch SDK, logs in, initializes a new workflow, and registers it to the platform. Requires Docker, Latch account credentials, and Python 3.8+.

```bash
# Install Latch SDK
uv pip install latch

# Login to Latch
latch login

# Initialize a new workflow
latch init my-workflow

# Register workflow to platform
latch register my-workflow
```

--------------------------------

### Setup Pi Agent from Source

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pi-agent/references/development.md

Clone the repository, install dependencies, and build the project. This is the initial step for development.

```bash
git clone https://github.com/earendil-works/pi-mono
cd pi-mono
npm install
npm run build
```

--------------------------------

### PyTorch Lightning DataModule: setup() Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pytorch-lightning/references/data_module.md

Use `setup` to create datasets and apply transforms. This method runs on every process in distributed training and is used for setting state like datasets.

```python
def setup(self, stage):
    if stage == 'fit':
        full_dataset = MyDataset("data/processed/")
        self.train_dataset, self.val_dataset = random_split(
            full_dataset, [0.8, 0.2]
        )

    if stage == 'test':
        self.test_dataset = MyDataset("data/processed/test/")

    if stage == 'predict':
        self.predict_dataset = MyDataset("data/processed/predict/")
```

--------------------------------

### Development Setup Commands

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/markdown-mermaid-writing/templates/project_documentation.md

Commands to set up the development environment, including cloning the repository, installing dependencies, and running tests and linters.

```bash
# Fork and clone
git clone https://github.com/[your-fork]/[repo].git

# Install with dev dependencies
[package-manager] install --dev

# Run tests
[package-manager] test

# Run linter
[package-manager] run lint
```

--------------------------------

### Vectorization Setup with pufferlib.make

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pufferlib/SKILL.md

Demonstrates automatic vectorization setup using pufferlib.make. Shows example performance benchmarks for different environment types.

```python
import pufferlib

# Automatic vectorization
env = pufferlib.make('environment_name', num_envs=256, num_workers=8)

# Performance benchmarks:
# - Pure Python envs: 100k-500k SPS
# - C-based envs: 100M+ SPS
# - With training: 400k-4M total SPS
```

--------------------------------

### Manual Multi-Node Setup - Node 0 (Master)

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pytorch-lightning/references/distributed_training.md

Command to start the training script on the master node for a manual multi-node setup. Includes arguments for total nodes, rank, master address, and port.

```bash
python train.py --num_nodes=2 --node_rank=0 --master_addr=192.168.1.1 --master_port=12345
```

--------------------------------

### Astropy Quick Start Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/astropy/SKILL.md

Demonstrates basic usage of astropy.units, astropy.coordinates, astropy.time, astropy.io.fits, astropy.table, and astropy.cosmology.

```python
import astropy.units as u
from astropy.coordinates import SkyCoord
from astropy.time import Time
from astropy.io import fits
from astropy.table import Table
from astropy.cosmology import Planck18

# Units and quantities
distance = 100 * u.pc
distance_km = distance.to(u.km)

# Coordinates
coord = SkyCoord(ra=10.5*u.degree, dec=41.2*u.degree, frame='icrs')
coord_galactic = coord.galactic

# Time
t = Time('2023-01-15 12:30:00')
jd = t.jd  # Julian Date

# FITS files
data = fits.getdata('image.fits')
header = fits.getheader('image.fits')

# Tables
table = Table.read('catalog.fits')

# Cosmology
d_L = Planck18.luminosity_distance(z=1.0)
```

--------------------------------

### gget setup

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/gget/references/module_reference.md

Install or download dependencies for specified gget modules.

```APIDOC
## gget setup

### Description
Install/download dependencies for modules.

### Method
CLI command

### Endpoint
`gget setup`

### Parameters
#### Path Parameters
None

#### Query Parameters
- **module** (str) - Required - Module name (e.g., alphafold, cellxgene, elm, gpt)
- **-o/--out** (str) - Optional - Output folder (elm only)
- **-q/--quiet** (flag) - Optional - Suppress progress (default: False)

### Request Example
```bash
gget setup --module gpt
```

### Response
#### Success Response (200)
- None (installs dependencies).

#### Response Example
```text
Setup for module 'gpt' completed.
```
```

--------------------------------

### Configure Environment Variables

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/markdown-mermaid-writing/templates/project_documentation.md

Copy the example environment file and edit it with your specific configuration values before starting the application.

```bash
cp .env.example .env
# Edit .env with your values
```

--------------------------------

### Install Vaex Minimal Components

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/vaex/SKILL.md

Install only the necessary Vaex sub-packages for a minimal setup, choosing specific components based on your needs.

```bash
uv pip install vaex-core vaex-viz vaex-hdf5 vaex-ml
```

--------------------------------

### gget setup - Install Dependencies

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/gget/SKILL.md

Installs or downloads third-party dependencies for specific gget modules. Supports Python dependencies via uv pip or pip.

```APIDOC
## gget setup - Install Dependencies

### Description
Installs or downloads third-party dependencies for specific gget modules. Supports Python dependencies via uv pip or pip.

### Parameters
#### CLI Parameters
- `module`: Module name requiring dependency installation
- `-o/--out`: Output folder path (elm module only)

#### Python Parameters
- `module`: Module name requiring dependency installation
- `out`: Output folder path (elm module only)

### Modules Requiring Setup
- `alphafold`: Downloads ~4GB of model parameters
- `cellxgene`: Installs cellxgene-census (may require Python 3.9/3.10 if the latest Python is unsupported)
- `elm`: Downloads local ELM database
- `gpt`: Installs/configures OpenAI integration dependencies

### Request Example
```bash
# Setup AlphaFold
gget setup alphafold

# Setup ELM with custom directory
gget setup elm -o /path/to/elm_data
```

```python
# Python
gget.setup("alphafold")
```
```

--------------------------------

### Manual Multi-Node Setup - Node 1

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pytorch-lightning/references/distributed_training.md

Command to start the training script on a worker node for a manual multi-node setup. Includes arguments for total nodes, rank, master address, and port.

```bash
python train.py --num_nodes=2 --node_rank=1 --master_addr=192.168.1.1 --master_port=12345
```

--------------------------------

### Complete PyLabRobot Visualization Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pylabrobot/references/visualization.md

A comprehensive example demonstrating the setup and execution of a liquid handling protocol with real-time visualization. This includes initializing the visualizer, liquid handler, resources, setting initial liquids, and performing transfers.

```python
from pylabrobot.liquid_handling import LiquidHandler
from pylabrobot.liquid_handling.backends.simulation import ChatterboxBackend
from pylabrobot.resources import (
    STARLetDeck,
    TIP_CAR_480_A00,
    Cos_96_DW_1mL,
    set_tip_tracking,
    set_volume_tracking
)
from pylabrobot.visualizer import Visualizer

# Enable tracking
set_tip_tracking(True)
set_volume_tracking(True)

# Create visualizer
vis = Visualizer()
await vis.start()

# Create liquid handler
lh = LiquidHandler(
    backend=ChatterboxBackend(num_channels=8),
    deck=STARLetDeck()
)
lh.visualizer = vis
await lh.setup()

# Define resources
tip_rack = TIP_CAR_480_A00(name="tips")
source_plate = Cos_96_DW_1mL(name="source")
dest_plate = Cos_96_DW_1mL(name="dest")

# Assign to deck
lh.deck.assign_child_resource(tip_rack, rails=1)
lh.deck.assign_child_resource(source_plate, rails=10)
lh.deck.assign_child_resource(dest_plate, rails=15)

# Set initial volumes
for well in source_plate.children:
    well.tracker.set_liquids([("sample", 200)])

# Execute protocol with visualization
await lh.pick_up_tips(tip_rack["A1:H1"])
await lh.transfer(
    source_plate["A1:H12"],
    dest_plate["A1:H12"],
    vols=100
)
await lh.drop_tips()

# Keep visualizer open to inspect final state
input("Press Enter to close visualizer...")

# Cleanup
await lh.stop()
await vis.stop()
```

--------------------------------

### Clone HypoGeniC Datasets

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/hypogenic/SKILL.md

Clone the HypoGeniC datasets repository for data-driven examples. This is useful for getting started with HypoGeniC's data-only functionalities.

```bash
# HypoGeniC examples (data-driven only)
git clone https://github.com/ChicagoHAI/HypoGeniC-datasets.git ./data
```

--------------------------------

### Benchling SDK Import Paths

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/benchling-integration/references/sdk_reference.md

Shows the preferred and an alternative valid import path for the Benchling SDK client. The preferred method is documented in the getting started guide.

```python
# Preferred (documented in getting started guide)
from benchling_sdk.benchling import Benchling

# Also valid in benchling-sdk 1.25+
from benchling_sdk import Benchling
```

--------------------------------

### API GET Response Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/markdown-mermaid-writing/templates/project_documentation.md

This is an example of a successful JSON response for a GET request to an API resource, including data and metadata.

```json
{
  "data": [
    {
      "id": "uuid",
      "name": "Example",
      "created_at": "2026-01-15T10:30:00Z"
    }
  ],
  "meta": {
    "total": 42,
    "limit": 20,
    "offset": 0
  }
}
```

--------------------------------

### Setup Plugin Distribution with setuptools

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/markitdown/references/api_reference.md

Configure setup.py to define package metadata and register the plugin's entry point for MarkItDown.

```python
from setuptools import setup

setup(
    name="markitdown-my-plugin",
    version="0.1.0",
    packages=["my_plugin"],
    entry_points={
        "markitdown.plugins": [
            "my_plugin = my_plugin.converter:MyConverter",
        ],
    },
)
```

--------------------------------

### Install and Start MATLAB Engine for Python

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/matlab/references/python-integration.md

Install the MATLAB Engine API for Python by running `setup.py` from the MATLAB installation directory. Start a MATLAB engine session using `matlab.engine.start_matlab()` or connect to an existing shared session.

```python
# Install MATLAB Engine API for Python
# From MATLAB: cd(fullfile(matlabroot,'extern','engines','python'))
# Then: python setup.py install

import matlab.engine

# Start MATLAB engine
eng = matlab.engine.start_matlab()

# Or connect to shared session (MATLAB: matlab.engine.shareEngine)
eng = matlab.engine.connect_matlab()

# List available sessions
matlab.engine.find_matlab()
```

--------------------------------

### Create and List Projects

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/latchbio-integration/references/data-management.md

Demonstrates how to create a new project with a name and description, and how to list all existing projects.

```python
from latch.registry.project import Project

# Get or create a project
project = Project.create(
    name="RNA-seq Analysis",
    description="Bulk RNA-seq experiments"
)

# List existing projects
all_projects = Project.list()

# Get project by ID
project = Project.get(project_id="proj_123")
```

--------------------------------

### BLS API Get Single Series Data (GET Request Example)

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/bls.md

Example of a GET request to retrieve data for a single series ID. This format is convenient for quick lookups and includes optional parameters like API key and date range.

```http
https://api.bls.gov/publicAPI/v2/timeseries/data/CUUR0000SA0?registrationkey=YOUR_KEY&startyear=2022&endyear=2024
```

--------------------------------

### Unpaywall DOI Lookup Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/paper-lookup/references/unpaywall.md

An example of a GET request to the Unpaywall DOI lookup endpoint.

```text
https://api.unpaywall.org/v2/10.1038/nature12373?email=you@example.com
```

--------------------------------

### Get Work by DOI Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/paper-lookup/references/crossref.md

An example of how to retrieve a work's metadata using its DOI.

```http
https://api.crossref.org/works/10.1038%2Fnature12373?mailto=you@example.com
```

--------------------------------

### Autoskill CLI: Basic Workflow Examples

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/autoskill/SKILL.md

Illustrates the basic commands for operating the autoskill CLI, including doctor, run, and promote.

```bash
python scripts/autoskill.py doctor   --config config.yaml --skills-dir ../
python scripts/autoskill.py run      --start ... --end ... --config config.yaml
python scripts/autoskill.py promote  --proposed ~/.autoskill/proposed/<ts> --skills-dir ../ --name <skill>
```

--------------------------------

### Initialize Benchling SDK with API Key

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/benchling-integration/references/authentication.md

Demonstrates how to initialize the Benchling SDK using an API key and tenant URL retrieved from environment variables. Ensure that BENCHLING_API_KEY and BENCHLING_TENANT_URL are set before running.

```python
import os
from benchling_sdk.benchling import Benchling
from benchling_sdk.auth.api_key_auth import ApiKeyAuth

api_key = os.environ.get("BENCHLING_API_KEY")
tenant_url = os.environ.get("BENCHLING_TENANT_URL")

if not api_key or not tenant_url:
    raise ValueError("Set BENCHLING_API_KEY and BENCHLING_TENANT_URL")

benchling = Benchling(
    url=tenant_url,
    auth_method=ApiKeyAuth(api_key),
)
```

--------------------------------

### Get a Study Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/gwas-catalog.md

Example of how to retrieve details for a specific study using its accession ID.

```text
https://www.ebi.ac.uk/gwas/rest/api/studies/GCST001633
```

--------------------------------

### Install NVIDIA Warp

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/optimize-for-gpu/references/warp.md

Install the Warp library using uv. The CUDA 12 runtime is the most common option. You can also install with example dependencies.

```bash
uv add warp-lang              # CUDA 12 runtime (most common)
# uv add warp-lang[examples]  # Includes USD and example dependencies
```

--------------------------------

### Complete Function Configuration Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/modal/references/resources.md

An example demonstrating the configuration of multiple resource parameters for a Modal function, including CPU, memory, GPU, ephemeral disk, timeout, and container scaling.

```python
@app.function(
    cpu=8.0,              # 8 physical cores
    memory=32768,         # 32 GiB
    gpu="L40S",           # L40S GPU
    ephemeral_disk=204800, # 200 GiB temp disk
    timeout=7200,         # 2 hours
    max_containers=50,
    min_containers=1,
)
def full_pipeline(data_path: str):
    ...
```

--------------------------------

### Install, Create, and Test nf-core Subworkflows

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/nextflow/references/nf-core-tools.md

Demonstrates how to install a pre-existing subworkflow, create a new one, and run its associated tests.

```bash
nf-core subworkflows install bam_sort_stats_samtools
nf-core subworkflows create align_bwa
nf-core subworkflows test align_bwa
```

--------------------------------

### Get Plasmid Details Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/addgene.md

Example of how to retrieve details for a specific plasmid using its Addgene ID.

```http
GET https://www.addgene.org/api/plasmids/12260/
Authorization: Token YOUR_KEY
```

--------------------------------

### BLS API Latest Data (GET Request Example)

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/bls.md

Example of a GET request to retrieve the latest available data for a series ID without specifying a date range. This will return data for the most recent three years.

```http
https://api.bls.gov/publicAPI/v2/timeseries/data/LNS14000000?registrationkey=YOUR_KEY
```

--------------------------------

### Example: Use ODE Solver and Different Noise Schedules

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/diffdock/references/parameters_reference.md

Illustrates enabling the ODE solver instead of the default SDE and using different noise schedules per component for alternative sampling.

```bash
python -m inference --ode --different_schedules
```

--------------------------------

### Install Benchling SDK

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/benchling-integration/references/sdk_reference.md

Installs the stable release or preview builds of the Benchling SDK using uv pip.

```bash
# Stable release (recommended)
uv pip install "benchling-sdk==1.25.0"
```

```bash
# Preview builds — alpha functionality, not for production
uv pip install "benchling-sdk" --prerelease allow
```

--------------------------------

### arXiv API Query Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/paper-lookup/references/arxiv.md

Example of a GET request to the arXiv API with search query and pagination parameters.

```http
GET https://export.arxiv.org/api/query?search_query={query}&start={n}&max_results={n}
```

--------------------------------

### REST API Search Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/opentargets.md

Example of a REST GET request to search for 'TP53' and retrieve up to 5 results.

```http
https://api.platform.opentargets.org/api/v4/search?q=TP53&size=5
```

--------------------------------

### Get GO Term Details Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/quickgo.md

Example of how to retrieve details for a specific GO term using its ID.

```text
https://www.ebi.ac.uk/QuickGO/services/ontology/go/terms/GO:0003723
```

--------------------------------

### Install Adaptyv SDK from GitHub

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/adaptyv/SKILL.md

Install the beta version of the adaptyv-sdk from GitHub using uv pip. Ensure you are using version 0.1.0 or later.

```bash
uv pip install "git+https://github.com/adaptyvbio/adaptyv-sdk.git"
```

--------------------------------

### Get Associations for a SNP Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/gwas-catalog.md

Example of how to retrieve all associations linked to a specific Single Nucleotide Polymorphism (SNP) by its rsID.

```text
https://www.ebi.ac.uk/gwas/rest/api/singleNucleotidePolymorphisms/rs7329174/associations
```

--------------------------------

### Install gget Dependencies with gget setup (CLI)

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/gget/SKILL.md

Use the gget setup command to install third-party dependencies for specific gget modules like AlphaFold, cellxgene, ELM, or GPT. Specify an output directory for the ELM module if needed.

```bash
gget setup alphafold
```

```bash
gget setup elm -o /path/to/elm_data
```

--------------------------------

### Gondolin Extension Setup and Execution

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pi-agent/references/containerization.md

Steps to install and run the Gondolin extension. This requires Node.js >= 23.6.0 and QEMU. The extension mounts the host cwd at /workspace in the VM.

```bash
cp -R packages/coding-agent/examples/extensions/gondolin ~/.pi/agent/extensions/gondolin
cd ~/.pi/agent/extensions/gondolin
npm install --ignore-scripts
cd /path/to/project
pi -e ~/.pi/agent/extensions/gondolin
```

--------------------------------

### Install numpy with version pinning

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/SECURITY.md

This example demonstrates installing the numpy package with a specific version (1.24.0). Version pinning is a best practice to ensure consistent and secure installations.

```bash
subprocess.check_call(['pip', 'install', 'numpy==1.24.0'])
```

--------------------------------

### Install PyTorch and PyG Core

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/torch-geometric/SKILL.md

Installs PyTorch and the core PyTorch Geometric library using uv pip. Ensure PyTorch is installed first, matching your CUDA/CPU setup.

```bash
# 1. Install PyTorch first (match your CUDA/CPU setup — see https://pytorch.org/get-started/locally/)
uv pip install torch

# 2. Core PyG (no extension wheels required for basic usage)
uv pip install torch_geometric
```

--------------------------------

### Install Qiskit with Visualization and Matplotlib

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/qiskit/references/setup.md

Install Qiskit along with visualization capabilities and Matplotlib for plotting.

```bash
uv pip install "qiskit[visualization]" matplotlib
```

--------------------------------

### List GDC Projects Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/tcga-gdc.md

Example GET request to list cancer projects, retrieving specific fields and limiting the results.

```http
https://api.gdc.cancer.gov/projects?size=5&fields=project_id,name,primary_site
```

--------------------------------

### Example: Set Batch Size and Enable Progress Bar

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/diffdock/references/parameters_reference.md

Demonstrates setting the batch size for processing and enabling the tqdm progress bar for monitoring long-running jobs.

```bash
python -m inference --batch_size 100 --tqdm
```

--------------------------------

### Fetch Gene by Hugo Symbol Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/cbioportal.md

Example GET request to retrieve information about a specific gene using its Hugo symbol.

```http
GET https://www.cbioportal.org/api/genes/TP53
```

--------------------------------

### Python: Submit a Binding Screen Step by Step

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/adaptyv/SKILL.md

This workflow demonstrates the step-by-step process of submitting a binding screen experiment using the Adaptyv client. It covers finding a target, estimating costs, creating the experiment, submitting it for review, and retrieving results.

```python
# 1. Find a target
targets = client.targets.list(search="EGFR", selfservice_only=True)
target_id = targets.items[0].id

# 2. Preview cost
estimate = client.experiments.cost_estimate({
    "experiment_spec": {
        "experiment_type": "screening",
        "method": "bli",
        "target_id": target_id,
        "sequences": {"seq1": "EVQLVESGGGLVQ...", "seq2": "MKVLVAG..."},
        "n_replicates": 3
    }
})

# 3. Create experiment (starts as Draft)
exp = client.experiments.create({
    "name": "EGFR binder screen batch 1",
    "experiment_spec": {
        "experiment_type": "screening",
        "method": "bli",
        "target_id": target_id,
        "sequences": {"seq1": "EVQLVESGGGLVQ...", "seq2": "MKVLVAG..."},
        "n_replicates": 3
    }
})

# 4. Submit for review
client.experiments.submit(exp.experiment_id)

# 5. Poll or use webhooks until Done
# 6. Retrieve results
results = client.experiments.get_results(exp.experiment_id)
```

--------------------------------

### Example Intraday Stock Time Series Request

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/alphavantage.md

An example GET request for fetching 5-minute intraday stock data for AAPL.

```url
https://www.alphavantage.co/query?function=TIME_SERIES_INTRADAY&symbol=AAPL&interval=5min&apikey=YOUR_KEY
```

--------------------------------

### Basic Retrosynthesis Pipeline Setup

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/torchdrug/references/retrosynthesis.md

Demonstrates loading a dataset and initializing models and tasks for retrosynthesis, including center identification and synthon completion.

```python
from torchdrug import datasets, models, tasks

# Load dataset
dataset = datasets.USPTO50k("~/retro-datasets/")

# For center identification
model_center = models.RGCN(
    input_dim=dataset.node_feature_dim,
    num_relation=dataset.num_bond_type,
    hidden_dims=[256, 256, 256]
)

task_center = tasks.CenterIdentification(
    model_center,
    top_k=3  # Consider top 3 reaction centers
)

# For synthon completion
model_synthon = models.GIN(
    input_dim=dataset.node_feature_dim,
    hidden_dims=[256, 256, 256]
)

task_synthon = tasks.SynthonCompletion(
    model_synthon,
    center_topk=3,  # Use top 3 from center identification
    num_synthon_beam=5  # Beam search for synthon generation
)

# End-to-end
task_retro = tasks.Retrosynthesis(
    model=model_center,
    synthon_model=model_synthon,
    center_topk=5,
    num_synthon_beam=10
)
```

--------------------------------

### Initialize and Query ModelStore

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/molfeat/references/examples.md

Demonstrates how to initialize the ModelStore, list available models, search for specific models by name, and retrieve model information. Use this to explore and understand the pretrained models available.

```python
from molfeat.store.modelstore import ModelStore

# Initialize model store
store = ModelStore()

# List all available models
print(f"Total available models: {len(store.available_models)}")

# Search for specific models
chemberta_models = store.search(name="ChemBERTa")
for model in chemberta_models:
    print(f"- {model.name}: {model.description}")

# Get model information
model_card = store.search(name="ChemBERTa-77M-MLM")[0]
print(f"Model: {model_card.name}")
print(f"Version: {model_card.version}")
print(f"Authors: {model_card.authors}")

# View usage instructions
model_card.usage()

# Load model directly
transformer = store.load("ChemBERTa-77M-MLM")
```

--------------------------------

### Install PyHealth with uv

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pyhealth/SKILL.md

Use uv for environment management to install PyHealth and its dependencies. This ensures a faster and reproducible environment setup.

```bash
# Create a project with the right Python
uv init my-pyhealth-project
cd my-pyhealth-project
uv python pin 3.12

# Add PyHealth (this also pulls in PyTorch and friends)
uv add pyhealth

# Run scripts inside the env
uv run python train.py
```

--------------------------------

### Client Pattern Examples

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/adaptyv/SKILL.md

Examples demonstrating the usage of the `FoundryClient` for interacting with the Adaptyv API.

```APIDOC
## Client Pattern

This pattern uses the `FoundryClient` to interact with the Adaptyv API.

### Initialization

```python
import os
from adaptyv import FoundryClient

client = FoundryClient(
    api_key=os.environ["ADAPTYV_API_KEY"],
    base_url=os.environ.get(
        "ADAPTYV_API_URL",
        "https://foundry-api-public.adaptyvbio.com/api/v1",
    ),
)
```

### Browsing Targets

```python
# Browse targets
targets = client.targets.list(search="EGFR", selfservice_only=True)
```

### Estimating Experiment Cost

```python
estimate = client.experiments.cost_estimate({
    "experiment_spec": {
        "experiment_type": "screening",
        "method": "bli",
        "target_id": "target-uuid",
        "sequences": {"seq1": "EVQLVESGGGLVQ..."},
        "n_replicates": 3
    }
})
```

### Creating and Submitting Experiments

```python
# Create and submit
exp = client.experiments.create({...})
client.experiments.submit(exp.experiment_id)
```

### Retrieving Experiment Results

```python
# Later: retrieve results
results = client.experiments.get_results(exp.experiment_id)
```

### `FoundryClient` Methods

- **`targets.list(search: str, selfservice_only: bool)`**: Lists available targets.
- **`experiments.cost_estimate(experiment_spec: dict)`**: Estimates the cost of an experiment.
- **`experiments.create(experiment_data: dict)`**: Creates a new experiment.
- **`experiments.submit(experiment_id: str)`**: Submits an experiment for execution.
- **`experiments.get_results(experiment_id: str)`**: Retrieves the results of an experiment.
```

--------------------------------

### Basic FSDP Trainer Setup

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pytorch-lightning/references/distributed_training.md

Initialize a PyTorch Lightning Trainer with the FSDP strategy for distributed training across multiple GPUs.

```python
trainer = L.Trainer(
    strategy="fsdp",
    accelerator="gpu",
    devices=4
)
```

--------------------------------

### Install pi-interview and GlimpseUI

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pi-agent/references/pi-interview.md

Install the pi-interview package and optionally the glimpseui package for native macOS windows. This is the initial setup step.

```bash
pi install npm:pi-interview
pi install npm:glimpseui   # optional: native macOS windows (browser fallback otherwise)
```

--------------------------------

### Install Qiskit and Visualization Tools

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/qiskit/SKILL.md

Installs the Qiskit library and additional packages for visualization and plotting. Use this command to set up your environment for Qiskit development.

```bash
uv pip install qiskit
uv pip install "qiskit[visualization]" matplotlib
```

--------------------------------

### Install Pi Coding Agent

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pi-agent/references/overview.md

Installs the Pi coding agent globally using npm. Use this for a quick setup on most systems.

```bash
npm install -g --ignore-scripts @earendil-works/pi-coding-agent
pi
```

--------------------------------

### PMC ID Converter API Example Request

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/paper-lookup/references/pmc.md

An example request to the ID Converter API to get JSON output for a given PMCID.

```http
https://pmc.ncbi.nlm.nih.gov/tools/idconv/api/v1/articles/?ids=PMC7029759&format=json
```

--------------------------------

### Install Qiskit with uv

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/qiskit/references/setup.md

Use uv to install the base Qiskit package.

```bash
uv pip install qiskit
```

--------------------------------

### Install Unpinned Preview Benchling SDK

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/SECURITY.md

This command installs a preview/alpha build of the benchling-sdk without a version pin, which could lead to the installation of an untested or compromised package. Use with caution and consider pinning to a specific version.

```bash
uv pip install "benchling-sdk" --prerelease allow
```

--------------------------------

### Example: Get DNA Sequence

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/ucsc-genome.md

An example API call to retrieve DNA sequence for a specific region on chromosome 7 in the hg38 genome.

```bash
GET https://api.genome.ucsc.edu/getData/sequence?genome=hg38&chrom=chr7&start=117119148&end=117119178
```

--------------------------------

### Fetch Clinical Data Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/cbioportal.md

Example GET request to retrieve patient clinical data for a study, filtering by a specific attribute and projection.

```http
GET https://www.cbioportal.org/api/studies/brca_tcga/clinical-data?clinicalDataType=PATIENT&attributeId=OS_STATUS&projection=SUMMARY
```

--------------------------------

### List Molecular Profiles in a Study Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/cbioportal.md

Example GET request to list all molecular profiles available within a specific cancer study.

```http
GET https://www.cbioportal.org/api/studies/brca_tcga/molecular-profiles
```

--------------------------------

### Quick Setup for Vectorized Environments

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/stable-baselines3/SKILL.md

Sets up multiple environment instances in parallel using `make_vec_env` for faster training. This example uses `SubprocVecEnv` for parallel execution, suitable for compute-heavy environments.

```python
from stable_baselines3.common.env_util import make_vec_env
from stable_baselines3.common.vec_env import SubprocVecEnv

# Create 4 parallel environments
env = make_vec_env("CartPole-v1", n_envs=4, vec_env_cls=SubprocVecEnv)

model = PPO("MlpPolicy", env, verbose=1)
model.learn(total_timesteps=25000)
```

--------------------------------

### Install Optional MarkItDown Dependencies

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/markitdown/SKILL.md

Control the file format support for MarkItDown by installing specific optional dependencies using pip. This example shows how to install support for PDF, DOCX, and PPTX files.

```bash
# Install specific formats
pip install 'markitdown[pdf, docx, pptx]'
```

--------------------------------

### Initialize and Use ModelStore

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/molfeat/references/api_reference.md

Demonstrates initializing the ModelStore, listing available models, searching for a specific model, and loading it. Use this to discover and access pre-trained featurizers.

```python
from molfeat.store.modelstore import ModelStore

# Initialize store
store = ModelStore()

# List all available models
all_models = store.available_models
print(f"Found {len(all_models)} featurizers")

# Search for specific model
results = store.search(name="ChemBERTa-77M-MLM")
if results:
    model_card = results[0]

    # View usage information
    model_card.usage()

    # Load the model
    transformer = model_card.load()

# Direct loading
transformer = store.load("ChemBERTa-77M-MLM")
```

--------------------------------

### Install labarchives-py Wrapper

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/labarchive-integration/SKILL.md

Clone the repository, navigate to the directory, and install the Python wrapper using uv pip.

```bash
git clone https://github.com/mcmero/labarchives-py
cd labarchives-py
uv pip install .
```

--------------------------------

### Pin Dependency Installation

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/SECURITY.md

Pinning dependencies to specific versions prevents supply chain attacks. This example shows how to install 'omero-py' with a specific version.

```bash
uv pip install omero-py==5.18.0
```

--------------------------------

### Quick Extension Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pi-agent/references/extensions.md

A basic Pi Agent extension demonstrating event handling, tool registration, and command registration. Use this as a starting point for custom agent functionalities.

```typescript
import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
import { Type } from "typebox";

export default function (pi: ExtensionAPI) {
  pi.on("session_start", async (_event, ctx) => {
    ctx.ui.notify("Extension loaded", "info");
  });

  pi.on("tool_call", async (event, ctx) => {
    if (event.toolName === "bash" && event.input.command?.includes("rm -rf")) {
      const ok = await ctx.ui.confirm("Dangerous", "Allow rm -rf?");
      if (!ok) return { block: true, reason: "Blocked by user" };
    }
  });

  pi.registerTool({
    name: "greet",
    label: "Greet",
    description: "Greet someone by name",
    parameters: Type.Object({ name: Type.String() }),
    async execute(_toolCallId, params) {
      return { content: [{ type: "text", text: `Hello, ${params.name}!` }], details: {} };
    },
  });

  pi.registerCommand("hello", {
    description: "Say hello",
    handler: async (args, ctx) => ctx.ui.notify(`Hello ${args || "world"}`, "info"),
  });
}
```

--------------------------------

### Get Node Property (Specific)

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/datacommons.md

Retrieve a specific property value for a given node. For example, to get the name of a place, use '->name'.

```HTTP
https://api.datacommons.org/v2/node?key=YOUR_KEY&nodes=geoId/06&property=->name
```

--------------------------------

### Setup Inheco Incubator

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pylabrobot/references/material-handling.md

Initialize and set up an Inheco incubator. Requires importing TemperatureController and InhecoBackend.

```python
from pylabrobot.temperature_control import TemperatureController
from pylabrobot.temperature_control.inheco import InhecoBackend

# Create incubator
incubator = TemperatureController(
    name="incubator",
    backend=InhecoBackend(),
    size_x=156.0,
    size_y=156.0,
    size_z=50.0
)

await incubator.setup()
```

--------------------------------

### Example Search Works URL

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/paper-lookup/references/core.md

An example of a GET request to the search works endpoint, querying for 'CRISPR gene therapy' with a limit of 10 results.

```http
https://api.core.ac.uk/v3/search/works/?q=CRISPR+gene+therapy&limit=10
```

--------------------------------

### List Studies Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/database-lookup/references/cbioportal.md

Example GET requests to list studies, demonstrating projection and page size parameters, and fetching a specific study by its ID.

```http
GET https://www.cbioportal.org/api/studies?projection=SUMMARY&pageSize=10
GET https://www.cbioportal.org/api/studies/brca_tcga
```

--------------------------------

### Multi-Device Workflow Example

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/pylabrobot/references/analytical-equipment.md

Demonstrates coordinating a liquid handler, plate reader, and scale in a single asynchronous workflow. Ensure all devices are initialized before use and stopped properly afterwards.

```python
async def multi_device_workflow():
    """Coordinate liquid handler, plate reader, and scale"""

    # Initialize all devices
    lh = LiquidHandler(backend=STAR(), deck=STARLetDeck())
    pr = PlateReader(name="CLARIOstar", backend=CLARIOstarBackend())
    scale = Scale(name="scale", backend=MettlerToledoBackend())

    await lh.setup()
    await pr.setup()
    await scale.setup()

    try:
        # 1. Weigh reagent
        await scale.tare()
        # (place container on scale)
        reagent_weight = await scale.get_weight()

        # 2. Prepare samples with liquid handler
        await lh.pick_up_tips(tip_rack["A1:H1"])
        await lh.transfer(source["A1:H12"], dest["A1:H12"], vols=100)
        await lh.drop_tips()

        # 3. Read plate
        await pr.open()
        # (load plate)
        await pr.close()
        data = await pr.read_absorbance(wavelength=450)

        return {
            "reagent_weight": reagent_weight,
            "absorbance_data": data
        }

    finally:
        await lh.stop()
        await pr.stop()
        await scale.stop()
```

--------------------------------

### GeoPandas Quick Start

Source: https://github.com/k-dense-ai/scientific-agent-skills/blob/main/skills/geopandas/SKILL.md

A quick start guide to using GeoPandas for reading, exploring, plotting, reprojecting, calculating areas, and saving geospatial data.

```python
import geopandas as gpd

# Read spatial data
gdf = gpd.read_file("data.geojson")

# Basic exploration
print(gdf.head())
print(gdf.crs)
print(gdf.geometry.geom_type)

# Simple plot
gdf.plot()

# Reproject to different CRS
gdf_projected = gdf.to_crs("EPSG:3857")

# Calculate area (use projected CRS for accuracy)
gdf_projected['area'] = gdf_projected.geometry.area

# Save to file
gdf.to_file("output.gpkg")
```