mukul975/privacy-data-protection-skills

connecticut-ctdpa

- Connecticut Data Privacy Act (CTDPA) compliance.

Vedi sorgente

Come usare

  1. Copia il comando e controlla il pacchetto prima di installarlo nel tuo ambiente agente.
  2. Controlla strumenti, credenziali e ambienti richiesti nel documento originale.
  3. Descrivi input e risultato desiderato; confronta l’output con le fonti.
npx skills add https://github.com/mukul975/privacy-data-protection-skills --skill connecticut-ctdpa

Prima di iniziare

  • L’installazione non include abbonamenti, quote API o accesso ai file.
  • Stelle e installazioni indicano attività, non un test indipendente di qualità o sicurezza.
Documento Skill originale

Contenuto dal repository con titoli, esempi, codice, tabelle, link e immagini preservati.

Connecticut Data Privacy Act (CTDPA)

Overview

The Connecticut Data Privacy Act (CTDPA), codified as Conn. Gen. Stat. §42-515 through §42-525, was signed into law on May 10, 2022 (SB 6), and became effective July 1, 2023. Connecticut was the fifth state to enact comprehensive consumer privacy legislation. The CTDPA is notable for its explicit prohibition on dark patterns in obtaining consent, its bona fide loyalty program exemption, and its requirement to recognize universal opt-out mechanisms (effective January 1, 2025).

Applicability (§42-516)

The CTDPA applies to persons that conduct business in Connecticut or produce products or services targeted to Connecticut residents AND during the preceding calendar year:

  1. Controlled or processed personal data of at least 100,000 Connecticut consumers (excluding data processed solely for payment transactions); OR
  2. Controlled or processed personal data of at least 25,000 Connecticut consumers AND derived more than 25% of gross revenue from the sale of personal data.

Note: The 25% revenue threshold is lower than Virginia's 50% and mirrors no specific percentage in Colorado.

Exemptions: State/local government, GLBA institutions (entity-level), HIPAA covered entities/BAs (entity-level), nonprofits, higher education institutions, and data governed by certain federal statutes.

Liberty Commerce Inc. Assessment: Liberty Commerce Inc. processes personal data of approximately 87,000 Connecticut consumers. With 12% advertising revenue from data sale activities, it does not meet threshold (2). However, excluding payment-only transactions, the consumer count exceeds the threshold under a conservative reading. Liberty Commerce Inc. treats itself as subject to the CTDPA.

Consumer Rights (§42-520)

Six Consumer Rights

  1. Right to Access (§42-520(a)(1)): Confirm processing and access personal data
  2. Right to Correct (§42-520(a)(2)): Correct inaccuracies
  3. Right to Delete (§42-520(a)(3)): Delete personal data
  4. Right to Portability (§42-520(a)(4)): Obtain data in portable, readily usable format
  5. Right to Opt Out (§42-520(a)(5)): Opt out of:
  • Targeted advertising
  • Sale of personal data
  • Profiling in furtherance of decisions producing legal or similarly significant effects
  1. Right to Appeal (§42-520(f)): Appeal a denial of any rights request

Response Requirements (§42-520(b)-(e))

  • Respond within 45 days of receipt
  • Extension: up to 45 additional days with notice
  • At least one free response per 12 months per right
  • Appeal: controller must respond within 60 days

Dark Pattern Prohibition (§42-515(8), §42-518(a)(5))

The CTDPA explicitly defines and prohibits dark patterns in the context of obtaining consumer consent. A "dark pattern" is defined as:

"A user interface designed or manipulated with the substantial effect of subverting or impairing user autonomy, decision-making or choice, and includes, but is not limited to, any practice the Federal Trade Commission refers to as a 'dark pattern'."

Implications

  • Consent obtained through dark patterns is not valid consent under the CTDPA
  • Applies to consent for sensitive data processing (§42-518(a)(5))
  • Applies to consent for processing children's data
  • Any consent mechanism that uses confusing language, hidden options, visual manipulation, or other deceptive practices is invalid

Dark Pattern Examples (per FTC guidance and CTDPA legislative history)

PatternDescriptionCTDPA Impact
Confirm-shamingUsing guilt-inducing language for the opt-out optionConsent invalid
Hidden optionsMaking the reject/decline option visually less prominentConsent invalid
Forced actionRequiring consent to access unrelated featuresConsent invalid
Trick questionsUsing double negatives or confusing languageConsent invalid
MisdirectionVisual design that steers users toward consentConsent invalid
NaggingRepeated prompts after consumer declinesConsent invalid

Liberty Commerce Inc. Implementation: Liberty Commerce Inc. has conducted a dark pattern audit of all consent interfaces. Consent buttons use equal visual weight (same size, same font, same color prominence). "Accept" and "Decline" options are presented side by side. Language is affirmative and clear. No confirm-shaming language is used.

Bona Fide Loyalty Program Exemption (§42-517(c))

The CTDPA provides an exemption for bona fide loyalty, rewards, premium, discount, or club card programs. Personal data collected as part of such programs is exempt from the sale opt-out right if:

  1. The program is a bona fide loyalty program (genuine value provided to the consumer)
  2. The consumer voluntarily participates
  3. The controller clearly discloses the data collection in the loyalty program terms
  4. The data is used solely for the purpose of the loyalty program

Liberty Commerce Inc. Implementation: Liberty Commerce Inc. operates the "Liberty Rewards" program. The program terms clearly state that purchase history and preference data are used exclusively for rewards calculation, personalized offers within the program, and member benefits. This data is not sold or used for cross-context behavioral advertising outside the loyalty program scope.

Universal Opt-Out Mechanism (§42-520(a)(6))

Effective January 1, 2025, controllers must recognize universal opt-out mechanisms. Requirements parallel Colorado's framework:

  • Must recognize and process universal opt-out signals (including GPC)
  • Signal applies to targeted advertising and sale of personal data
  • No additional consumer verification required
  • No pop-up or interstitial questioning the signal
  • Apply at browser/device level for unauthenticated; account level for authenticated

Sensitive Data (§42-518(a)(5))

Categories

Same as Virginia VCDPA:

  1. Racial or ethnic origin
  2. Religious beliefs
  3. Mental or physical health condition or diagnosis
  4. Sexual orientation
  5. Citizenship or immigration status
  6. Genetic or biometric data for identification
  7. Personal data of a known child
  8. Precise geolocation data

Consent Requirement

Processing requires opt-in consent that is not obtained through a dark pattern.

Controller Obligations (§42-518)

  1. Purpose limitation: Limit processing to what is adequate, relevant, and reasonably necessary
  2. Data minimization: Do not collect more than reasonably necessary
  3. Data security: Appropriate technical and organizational measures
  4. Non-discrimination: No processing in violation of antidiscrimination laws
  5. Sensitive data consent: Obtain consent (not via dark pattern) before processing
  6. Privacy notice: Reasonably accessible and clear
  7. DPIAs: Conduct assessments for targeted advertising, sale, profiling, sensitive data, and high-risk processing

Enforcement (§42-524, §42-525)

  • Exclusive AG enforcement — no private right of action
  • 60-day cure period until December 31, 2024
  • After January 1, 2025: AG has discretion on cure period
  • Penalties: Treated as unfair trade practices under Conn. Gen. Stat. §42-110b (CUTPA)
  • Civil penalties: Up to $5,000 per willful violation under CUTPA

Key Regulatory References

  • Conn. Gen. Stat. §42-515 through §42-525 (CTDPA)
  • Conn. Gen. Stat. §42-110b (Connecticut Unfair Trade Practices Act — enforcement)
  • FTC Dark Patterns Report (September 2022) — referenced in CTDPA definition
  • Global Privacy Control Specification v1.0 — recognized universal opt-out mechanism

Correlati per attività

dallo stesso repository

Altri Skills

Tutti gli Skills
mukul975
Community

applying-privacy-design-patterns

- Systematic application of the eight privacy design patterns per Hoepman: minimize, hide, separate, abstract, inform, control, enforce, and demonstrate. Covers pattern selection methodology per processing activity, mapping to GDPR principles, and practical implementation guidance for privacy-by-design system architecture.

installazioni
1
GitHub Stars
315
Aggiornato
16 mar
mukul975
Community

brazil-lgpd

- Guides compliance with Brazil's Lei Geral de Proteção de Dados (LGPD, Lei 13.709/2018). Covers the 10 lawful bases under Art. 7, DPO appointment, ANPD enforcement, data subject rights under Arts. 17-22, and international transfer mechanisms. Keywords: LGPD, Brazil data protection, ANPD, lawful bases, data subject rights, international transfers.

installazioni
1
GitHub Stars
315
Aggiornato
16 mar
mukul975
Community

adequacy-assessment

- Guides assessment of third-country adequacy decisions under GDPR Article 45 for international data transfers. Covers the current EC adequacy decisions list, adequacy assessment criteria, partial adequacy handling, and monitoring of adequacy decision reviews. Keywords: adequacy decision, Article 45, third country, adequate protection, EC adequacy list.

installazioni
1
GitHub Stars
295
Aggiornato
16 mar
mukul975
Community

age-gating-services

- Implements age-gating mechanisms for online services to restrict access based on user age. Covers hard gates versus soft gates, neutral age prompts, re-verification triggers, circumvention prevention, and regulatory requirements under GDPR, COPPA, UK Online Safety Act, and DSA. Keywords: age gate, age restriction, neutral prompt, children, online services, access control.

installazioni
1
GitHub Stars
295
Aggiornato
16 mar