Skills
.Work. You
Rest

applying-differential-privacy

Community

Ship formal privacy guarantees, not guesswork.

Legal & Compliance#rdp#dp-sgd#membership inference#differential privacy#privacy accounting#epsilon budget
Authorrocklambros
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill helps you apply differential privacy correctly when a model, dashboard, or analytics release must protect individuals from membership inference, attribute disclosure, or reconstruction attacks.

Core Features & Use Cases

  • Threat-model first guidance: Distinguishes when differential privacy is the right tool versus access control, k-anonymity, or cryptographic privacy methods.
  • Budget and mechanism selection: Chooses appropriate epsilon, delta, and mechanism settings for Gaussian, Laplace, Exponential, or DP-SGD workloads.
  • Production-grade privacy accounting: Tracks composition with RDP or related accountants, warns about weak or sloppy guarantees, and supports release-ready DP statements.
  • Use Case: A clinical NLP team fine-tuning on patient summaries can use this Skill to justify a privacy budget, configure DP-SGD in PyTorch, and report a defensible guarantee.

Quick Start

Use the applying-differential-privacy skill to determine the right privacy budget, mechanism, and accounting method for your dataset or training run.

Dependency Matrix

Required Modules

None required

Components

references

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: applying-differential-privacy
Download link: https://github.com/rocklambros/rcs/archive/main.zip#applying-differential-privacy

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 537,000+ vetted skills library on demand.