WORK ITEM: Repo for Data Minimization and selective disclosure
-
Updated
Apr 12, 2022 - HTML
WORK ITEM: Repo for Data Minimization and selective disclosure
Local-first AI WAF, compliance gate, and cryptographic provenance engine for agentic workflows. Intercepts payloads, slashes token spend via Prose-Tax minimization, and mints tamper-evident audit chains on local silicon.
A simple data minimization and anonymization microservice wrapped around go-minimizer
Contains all the code used and submitted for the indiviual assignments of the Data Protection Technologies (https://coursecatalogue.uva.nl/xmlpages/page/2023-2024-en/search-course/course/110258) as part of the MSc in Computer Science (Big Data Engineering track).
Data minimization, pseudonymization, and anonymization helpers for Go
Organizational Cognitive Telemetry (OCT) — a system for capturing sanitized AI interaction telemetry to identify cognitive friction, knowledge gaps, and usage patterns across an organization, without storing sensitive content or individual identities.
Client-side metadata minimisation for Miden private notes.
Export minimal debugging reports from Playwright traces with an explicit omission manifest.
Reusable code-graph and Git co-change analysis for reviewing parallel changes. Local demos and a PR-coordination runtime. Source-available; hosted App suspended.
Source code of IJCB2024 paper "Controllable Privacy in Face Recognition: A Filter-based Approach"
Data Prism is an open-source privacy layer for Java/Spring teams putting LLM agents or MCP clients in front of internal APIs holding customer data. It pseudonymises personal data per privacy scope, refuses anything unclassified, and can keep a hash-chained audit trail.
A JavaScript client with a graphical user interface for Kalita, a text-to-speech software with a focus on data minimization and user privacy.
OpenID4VP DCQL request linter for EUDI Wallet verifiers: check every credential and claim alternative against a data-minimization policy. TypeScript library and CLI for SD-JWT VC and ISO mdoc (eIDAS 2.0).
privacy.md — a privacy boundary for personal AI agents, enforced before the data leaves your machine. AGENTS.md tells an agent how to work; privacy.md tells it what it may send.
🛡️ The consent firewall for AI agents. Your agent can use the world's public data — the world can't use yours. Field-level minimization, in-flow consent, LLM tokenization, prompt-injection quarantine & hash-chained privacy receipts. MCP server included.
What happens when privacy law meets skewed data? A reproducible 960-run computational experiment measuring the joint effects of data minimization and class imbalance on ML performance and fairness. PhD dissertation research.
Offline proof of concept: coarse categorical incident postcards from 512-channel classical residual frames, with no raw-data sharing.
Decide whether an activity event may be retained and produce a minimized record under an explicit consent/retention policy.
Large-scale privacy over-collection detector for voice-assistant apps, using LLM embeddings + anomaly detection to flag apps requesting more data than their functionality needs (ICSE 2024)
Open-source tool that maps which vendors receive which student data in U.S. K-12 school districts and measures each vendor's exposure.
To associate your repository with the data-minimization topic, visit your repo's landing page and select "manage topics."