PII Firewall

PII Firewall is an open-source, self-hosted tool that intercepts personally identifiable information before it reaches an LLM and restores original values in the response. It is presented for workflows such as clinical language processing, fintech support automation, and legal document review. Built-in profiles cover healthcare, finance, legal, and generic use, with custom profiles also supported. Detection options include regex, Presidio, GLiNER, Transformers, OpenAI Privacy Filter, hybrid configurations, and custom backends. For detected information, users can choose to keep it, pseudonymize it, redact it, generalize it, mask it, or hash it. The product says it supports more than 55 languages with automatic language detection and locale-specific patterns. Integration examples include OpenAI, Anthropic, LangChain, FastAPI, streaming transport, Ollama, LiteLLM, and arbitrary HTTP LLM endpoints. Tenant-scoped token mappings and a forget operation support deletion of vault mappings for a thread or case. Each call produces a trace with detected entity types, confidence scores, and applied replacements. The project uses the Apache 2.0 license and provides a Python package; its regex-only base installation has no ML dependencies.

PII Firewall focuses on intercepting and restoring personal information in LLM workflows, with multiple detection and disposition choices. Its open-source license and regex-only installation are useful considerations, while the maker notes that hashing is not part of built-in presets.

It is suited to teams handling sensitive information in healthcare, finance, or legal workflows that use LLMs. Python users seeking a self-hosted, open-source option may also find its deployment model relevant.