Kdense Ml Ai Selected
精选 K-Dense Scientific Agent Skills 的机器学习与人工智能工具包。用于 scikit-learn、PyTorch Lightning、Transformers、SHAP、time series ML、TimesFM、PyTorch Geometric、UMAP 等科研建模路线，并按 Auto-sci-research 的体育地理、城市暴露、空间公平、健康影响和论文解释边界进行封装。
Install
$ agentstack add skill-lzy599775-agent-auto-sci-skills-kdense-ml-ai-selected ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
K-Dense ML/AI Selected
This wrapper packages selected Machine Learning & AI skills from K-Dense-AI/scientific-agent-skills for Auto-sci-research.
Use it when a task needs technical playbooks for:
scikit-learnpytorch-lightningtransformersshaptimesfm-forecastingtorch-geometricumap-learn
Local Adaptation
Use these upstream skills with local research constraints:
- Define prediction target, population, spatial unit, temporal unit, and leakage risks before model choice.
- Prefer simple baselines before complex models.
- Separate predictive utility from causal interpretation.
- For exposure, accessibility, and spatial equity work, document spatial and temporal validation splits.
- Treat SHAP, feature importance, and embeddings as interpretation aids, not causal evidence.
For domain-specific guidance, also read:
../agent-auto-sci-ai-ml/references/k_dense_ml_ai_mapping.md../agent-auto-sci-ai-ml/references/ml_leakage_and_validation.md
Must Not Do
- Do not report only accuracy without calibration, uncertainty, and validation design.
- Do not mix training and test geographies or time periods without disclosure.
- Do not present model explanations as mechanisms unless the design supports that claim.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Lzy599775
- Source: Lzy599775/agent-auto-sci-skills
- License: MIT
- Homepage: https://lzy599775.github.io/agent-auto-sci-skills/
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.