A pure Python HTML5 parser that just works. No C extensions to compile. No system dependencies to install. No complex API to learn.

A Python library to sanitize/validate a string such as filenames/file-paths/etc.

μSBS: Static Binary Sanitization of Bare-metal Embedded Devices

A SOC Analyst's tool to automate the investigation & validation of possible Indicators of Compromise (IOCs) and perform various tasks including Phishing Email Analysis & Brand Monitoring to fasten the incident response.

A Python package to scrub PII

Strips boilerplate from Project Gutenberg text files

As a SOC Analyst, incident detection, analysis and mitigation is a rigorous task. The project aims at automating the detection, investigation & validation of possible Indicators of Compromise (IOCs) and perform various tasks including Phishing Email Analysis & Brand Monitoring to improve the potential security incident response.

A Django middleware to validate user file uploads and detect malicious content.

Modern data sanitization software in compliance with NIST SP 800-88 Guidelines

Open-source NVMe sanitization framework with 50 cycles and per-cycle Log Page 0x81 hardware confirmation. PDF Certificate of Destruction. Aligns with NIST SP 800-88 Rev.2 & IEEE 2883.1-2025. Source code: GPL v3. Specification: Proprietary.

MIRROR of https://codeberg.org/catseye/T-Rext : A command-line tool that attempts to rectify punctuation and spacing in (generated) text files

CloakMCP — local-first, deterministic sanitizer for secrets before LLMs

Capture and sanitize HAR (HTTP Archive) files with deep PII removal. Perfect for support diagnostics, security reviews, and test fixtures.

pathvalidate-cli is a command line interface for pathvalidate library. The tool can do sanitize/validate a string such as file-names/file-paths.

Open-source LLM sanitization proxy. Detects and replaces PII before it reaches AI providers, then restores it in responses. Self-hosted Docker, OpenAI-compatible gateway.

A privacy-first MCP server for bug bounty hunting with Claude Code. Sensitive data never reaches the AI.

Rule-based PII and secret redaction for Markdown documents — audit log, risk-level filtering, LLM pipeline ready

Corporate LLM gateway. Sanitizes traffic between developer Claude Code instances and Anthropic/OpenAI before it leaves the corp boundary.

Deterministic input sanitization for untrusted text — invisible characters, homoglyphs, and encoding tricks, handled before your code sees them. Zero dependencies, no ML. Python 3.12+.

Hostile-text normalization, inspection, and cleaning for LLM-adjacent systems.