verifying-training-data-erasure

Verify training-data erasure across ML pipeline stages and assemble DSR proof bundles.

2|Updated May 23, 2026

Search Tags:#gdpr #vector store #hipaa #dsr #erasure #model weights

Ready to run

npx skills add https://github.com/rocklambros/rcs --skill verifying-training-data-erasure

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: verifying-training-data-erasure
Source: https://github.com/rocklambros/rcs/tree/main/skills/security/verifying-training-data-erasure
Command: npx skills add https://github.com/rocklambros/rcs --skill verifying-training-data-erasure

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps teams verify and document that a subject’s data has been erased across the full ML pipeline after a right-to-erasure or right-to-amend request.

Core Features & Use Cases

Multi-stage erasure verification: Checks raw datasets, embeddings and vector stores, model weights, inference caches, and backups or replicas for lingering subject data.
Audit-ready proof bundles: Produces structured evidence with commands run, outputs, sign-offs, residual-risk statements, and retention dates for privacy review.
Compliance workflows: Supports GDPR, CCPA, HIPAA, and similar DSR cases where trained models or retrieval systems may still retain subject-derived information.

Quick Start

Ask this Skill to walk a specific subject through erasure verification and generate a DSR proof bundle for the relevant regulation.

Frequently Asked Questions about verifying-training-data-erasure

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

FAQPage Schema

How do I prove GDPR Article 17 erasure across ML model weights and vector stores?▼

ML data erasure verification checks raw datasets, embeddings, vector stores, model weights, inference caches, and backups for lingering subject data. It supports GDPR, CCPA, and HIPAA right-to-erasure requests by assembling structured DSR proof bundles for privacy review.

How do I document DSR erasure for trained models and retrieval systems?▼

DSR erasure documentation requires stage-by-stage deletion evidence across raw datasets, embeddings, model weights, caches, and backups. The proof bundle includes commands run, outputs, sign-offs, residual-risk statements, and retention dates for privacy review.

Does this data erasure verification support HIPAA and CCPA compliance workflows?▼

Data erasure verification supports HIPAA, CCPA, and GDPR compliance workflows for right-to-erasure requests. It verifies deletion across raw datasets, embeddings, model weights, caches, backups, and deployed fine-tunes to produce audit-ready proof bundles.

What is the best way to handle residual-risk when erasing training data from model weights?▼

Handling residual-risk when erasing training data from model weights requires explicit model-weight strategy selection and residual-risk acknowledgement. The verification process documents these statements in the DSR proof bundle alongside deletion evidence and retention dates.

Why does ML data erasure require checking inference caches and deployed fine-tunes?▼

ML data erasure requires checking inference caches and deployed fine-tunes because subject-derived information can persist in these stages after raw dataset deletion. Full verification ensures no lingering subject data remains across the entire ML pipeline.