---
title: "PyTDC (Therapeutics Data Commons) - Air Market"
description: "You are working on a therapeutics machine-learning problem and need data plus a fair way to measure against it. This reaches Therapeutics Data Commons through PyTDC: find what exists, load it, split…"
canonical: "https://airmarket.ai/app/pytdc"
---

[Air Market](/) / [Data & Analytics](/category/data) / PyTDC (Therapeutics Data Commons)

Skill by [K-Dense](/@k-dense-ai)

# PyTDC (Therapeutics Data Commons)

Find a therapeutics dataset and split it the way the task needs

## What this does

You are working on a therapeutics machine-learning problem and need data plus a fair way to measure against it. This reaches Therapeutics Data Commons through PyTDC: find what exists, load it, split it the way the task calls for, and score with the accepted metrics. Benchmark groups and molecular oracles are covered too.

### Top 3 things it does for you

- Browse what datasets are available
- Load an approved dataset
- Split data the way the task expects

## Information

**PyTDC (Therapeutics Data Commons) at a glance**

|  |  |
| --- | --- |
| Kind | Skill |
| Publisher | [K-Dense](/@k-dense-ai) |
| Category | [Data & Analytics](/category/data) |
| Works with | Agent Skills - an open standard supported by 40+ agents |
| Version | 1.1 |
| Size | 105 KB |
| Licence | MIT |
| Repository | [k-dense-ai/scientific-agent-skills](https://github.com/k-dense-ai/scientific-agent-skills/tree/81e316bc89f7fda176b092d2bceb959204710b0a/skills/pytdc) |
| GitHub stars | 34,239 (as of Nov 2026) |

## Security report

**Verified.** AIR scanned this exact version of PyTDC (Therapeutics Data Commons) and flagged nothing. AIR's verdict: **safe**.

**What PyTDC (Therapeutics Data Commons)'s check was run against**

|  |  |
| --- | --- |
| Commit | 81e316bc89f7fda176b092d2bceb959204710b0a |
| Content hash | 698f8db1c5ca14ce256868f43af43d30c6a845772ca820aa77d2bac8c40a3fa2 |
| Files read | [Browse the files at this exact commit](https://github.com/k-dense-ai/scientific-agent-skills/blob/81e316bc89f7fda176b092d2bceb959204710b0a/skills/pytdc/SKILL.md) |

A scan is named by the exact content it read, not by a version number. A different version is a different artifact, and nothing here applies to a version AIR has not scanned.

[Read the full security report for PyTDC (Therapeutics Data Commons)](/security/pytdc)

## About PyTDC (Therapeutics Data Commons)

Access to Therapeutics Data Commons through the PyTDC Python package.

It covers registry discovery for finding what exists, access to approved datasets, task-aware splits, and evaluator metrics for scoring.

Benchmark groups and bounded molecular-oracle workflows are included.

## What's inside

- SKILL.md - instructions
- references/datasets.md - reference
- references/oracles.md - reference
- references/sources.md - reference
- references/utilities.md - reference
- scripts/_common.py - script
- scripts/benchmark_evaluation.py - script
- scripts/cache_audit.py - script
- scripts/discover_metadata.py - script
- scripts/load_and_split_data.py - script
- scripts/molecular_generation.py - script

## Questions

### Is PyTDC (Therapeutics Data Commons) safe to install?

AIR scanned this exact version of PyTDC (Therapeutics Data Commons) and flagged nothing. AIR's verdict: safe. The verdict is named by the content that was read, at commit 81e316bc89f7, so it cannot drift onto a version nobody scanned. The full report is at https://airmarket.ai/security/pytdc.

### How do I install PyTDC (Therapeutics Data Commons)?

Open https://airmarket.ai/app/pytdc on Air Market and use the install action: Air Market hands Claude a prompt naming this exact skill, and Claude sets it up. Source, pinned to the exact commit this listing describes: https://github.com/k-dense-ai/scientific-agent-skills/tree/81e316bc89f7fda176b092d2bceb959204710b0a/skills/pytdc. Install from that link, not the branch.

## Related on Air Market

- [More Data & Analytics listings](/category/data)
- [Everything by K-Dense](/@k-dense-ai)

---

Canonical URL: <https://airmarket.ai/app/pytdc>