openai

openai / privacy-filter

Public

OpenAI Privacy Filter

396
23
100% credibility
Found Apr 23, 2026 at 396 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A local toolkit with an AI model to detect, mask, and redact personally identifiable information like names, emails, phones, and secrets in text for privacy-safe workflows.

How It Works

1
🔍 Discover privacy protector

You hear about a simple tool that spots and hides personal details like names, emails, and dates in any text to keep data safe.

2
📦 Set it up easily

With one quick step on your computer, the tool is ready to use without any hassle.

3
⬇️ It grabs its smarts

The tool automatically downloads everything it needs to understand and protect privacy.

4
✨ Paste your text

Type or drop in text with sensitive info, and watch it instantly find and mask what's private.

5
📁 Clean whole files

Feed it documents or logs, and get back versions where personal bits are safely hidden.

6
🔍 Check and tweak

See colorful previews of what was changed, and adjust settings for perfect results.

✅ Data stays private

Your texts are now clean and shareable, with all personal info protected effortlessly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 396 to 396 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is privacy-filter?

OpenAI Privacy Filter is a Python tool for detecting and redacting personally identifiable information (PII) like emails, phones, names, dates, addresses, and secrets in text. Run it locally via CLI with `opf "Alice was born on 1990-01-02."` for instant masking, pipes for pipelines, or files for batch jobs; it also offers eval on labeled data and finetuning your datasets. Like an hp elitebook privacy filter for text, it processes long contexts up to 128k tokens without chunking, outputting redacted text or JSON spans.

Why is it gaining traction?

It stands out as an openai github alternative to cloud services like openai github api or openai privacy api, running fully on-premises on laptops or browsers with just 1.5B params (50M active). Developers hook on the tunable precision/recall via operating points, Apache 2.0 license for commercial use, and seamless GPU/CPU support—no vendor lock-in or data leaks. Compared to openai github copilot or whisper integrations, it's purpose-built for high-throughput sanitization.

Who should use this?

Data engineers scrubbing logs before analytics, ML teams building openai github python pipelines needing PII guards, or compliance folks handling privacy policy deutsch requests. Ideal for privacy filter dell laptop workflows or openai privacy portal redaction in customer support transcripts.

Verdict

Grab it if you need local PII redaction—solid CLI, API, and docs make it production-ready despite 396 stars and 1.0% credibility score signaling early maturity. Finetune for your domain, but pair with human review for high-stakes use; test coverage looks sparse but Apache license lowers risk.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.