Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/responsible-ai-guide
Resources for trustworthy, fair, and ethical AI research
Passed 24 of 25 test cases with the skill, and 23 of 25 without it.
1 fixed23 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
A test case the skill fixed. Same model, same prompt, with and without the skill.
I'm auditing a healthcare prediction model for potential demographic bias. Could you show me a Python script using standard evaluation libraries to compute accuracy and recall metrics across demographic subgroups and calculate the maximum difference between groups?
| Checked | Without skill | With skill |
|---|---|---|
| Uses MetricFrame from Fairlearn to aggregate performance metrics across sensitive feature groups. | Missed | Passed |
| Calculates the performance spread across groups using the difference method on the metric frame. | Missed | Passed |
The instructions your model reads when the skill loads.
A comprehensive collection of resources for building trustworthy, fair, and ethical AI systems. Covers fairness metrics, bias detection and mitigation, explainability methods, privacy-preserving techniques, robustness testing, and governance frameworks. Essential reading for researchers working on AI safety, alignment, and deploying models in high-stakes domains.
Responsible AI
├── Fairness
│ ├── Bias detection (data, model, outcome)
│ ├── Fairness metrics (demographic parity, equalized odds)
│ ├── Bias mitigation (pre/in/post-processing)
│ └── Intersectional fairness
├── Explainability
│ ├── Feature attribution (SHAP, LIME, IG)
│ ├── Concept-based (TCAV, concept bottleneck)
│ ├── Counterfactual explanations
│ └── Mechanistic interpretability
├── Privacy
│ ├── Differential privacy
│ ├── Federated learning
│ ├── Membership inference attacks
│ └── Machine unlearning
├── Robustness
│ ├── Adversarial attacks/defenses
│ ├── Distribution shift
│ ├── Uncertainty quantification
│ └── Out-of-distribution detection
├── Safety & Alignment
│ ├── RLHF and preference learning
│ ├── Constitutional AI
│ ├── Red teaming
│ └── Guardrails and filters
└── Governance
├── Model cards
├── Datasheets for datasets
├── AI impact assessments
└── Regulatory compliance (EU AI Act)
test cases passed, vs 23 / 25 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.