skills.rest
kridaydave
kridaydaveCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ethos

Reviews research studies for ethical risk, consent scope, bias, and regulatory compliance.

1|Updated Jun 1, 2026
Search Tags:#data privacy#gdpr#informed consent#research integrity#irb#bias audit#ethics review
One-click install
npx skills add https://github.com/kridaydave/My_Skills --skill ethos-kridaydave
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ethos
Source: https://github.com/kridaydave/My_Skills/tree/main/ethos
Command: npx skills add https://github.com/kridaydave/My_Skills --skill ethos-kridaydave

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Research teams often discover ethical, privacy, or integrity problems only after publication — as retractions, regulatory penalties, or harmed subjects. This Skill reviews studies, datasets, and models before work proceeds, surfacing consent violations, bias, conflicts of interest, and dual-use risks while they are still cheap to fix. ## Core Features & Use Cases - Ethics & IRB Review: Audits study designs for harm risks, prepares IRB/ethics-board submissions, and drafts informed-consent materials covering purpose, withdrawal, and data handling. - Privacy & Consent Auditing: Checks consent scope chains, PII handling, de-identification, retention, and GDPR/HIPAA-style compliance for datasets and secondary data use. - Bias & Integrity Analysis: Audits samples, labels, and model outputs for disparate impact, and flags undisclosed conflicts of interest, plagiarism risk, and dual-use concerns. - Use Case: Before training a model on account-setup data, ask for a review — the Skill flags that setup consent does not cover model training, recommends fresh consent or anonymization, and routes the decision to a DPO. ## Quick Start Ask the assistant to review your study protocol or dataset plan for ethical risks, consent scope, bias, and privacy compliance before data collection begins.

Frequently Asked Questions about ethos

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review a research study for ethical risks before data collection?▼

Submit your study protocol or data source plan for review before collecting, scraping, or labeling anything. The review identifies who could be harmed, checks consent scope, maps applicable regulations, and ranks findings by severity with concrete fixes.

How to check if dataset consent covers a new use case?▼

Consent is scoped to its original purpose, so reusing data for model training or secondary analysis requires checking the actual consent chain. The review flags purpose-limitation violations and recommends fresh consent, anonymization, or exclusion.

Can this replace an IRB or legal review?▼

No. It prepares materials, flags risks, and advises, but explicitly identifies items requiring a real IRB, DPO, or legal counsel sign-off. It is a pre-review gate, not a substitute for institutional approval.

How do I audit a machine learning model for bias and fairness?▼

Provide the model's sample composition, labeling process, and output metrics. The audit checks per-group performance for disparate impact hidden by aggregate accuracy, and flags unrepresentative samples as integrity failures or deployment harms.

What are the limitations of an automated ethics review?▼

Jurisdiction-specific rules may require assumptions, and the review states them explicitly (e.g., general human-subjects plus GDPR-style standards). Final determinations on regulated sectors, animal subjects, or specific national laws need institutional experts.

Related Skills

Explore More Legal & Compliance Skills→
t46t46
14

ethics-review-support

Assess ethics approval needs and generate submission plans with checklists.

Community
Intermediate
lwokeraylwokeray

brand-content-review

Reviews marketing content against brand guidelines, claims, and channel requirements with severity-ranked findings.

Community
Advanced
andikarachmanandikarachman
14

model-card

Generate standardized model documentation in HuggingFace Model Card and NVIDIA Model Card++ formats.

Community
Advanced
abhishekchoudhariabhishekchoudhari
1

privacy-policy

Generate privacy policy drafts covering data collection and jurisdictional compliance.

Community
Advanced