datafog-python
Offline PII firewall for AI agents and LLM apps: fast local detection and redaction, Claude Code hook, LiteLLM guardrail. Zero network calls, one dependency.
- 80
- Python
- MIT
- 4 d ago
DataFog
the embeddable PII engine. keep PII out of your LLM calls.
Built from DataFog's public GitHub profile, United States of America. Not on Take My Repo yet, so there are no posts or updates here.
Offline PII firewall for AI agents and LLM apps: fast local detection and redaction, Claude Code hook, LiteLLM guardrail. Zero network calls, one dependency.
REST API for computing cross-modal similarity between images and text using the ColPaLI vision-language model
PII detection and privacy transformations for text and structured records. Built in Rust, with Python, Node.js, and browser SDKs.
Privacy Engineering for the Generative AI era made available through a REST API
Policy-first API and policy gate for AI/CLI workflows: detect sensitive data, decide allow/transform/deny, and emit auditable receipts.
Locally replaces detected personal data and selected credentials with reversible markers in Claude Code. See coverage and platform limits.
MCP server for local PII scanning and redaction. Check sensitive files and create redacted copies from Claude Code. Powered by DataFog Core.
Real lance-context training exports reviewed with DataFog Core, with a local UI and reproducible synthetic data.
Shared DataFog development environments, verification adapters, and pilot workspaces
OpenClaw plugin for PII detection & custom entity redaction powered by DataFog