Summary
noirdoc launched an open-source Claude Code plugin and companion API proxy that pseudonymize names, emails, IBANs, and other sensitive fields before model calls are made. The product keeps reversible mappings local and extends the same approach to OpenAI, Anthropic, Azure, and OpenAI-compatible endpoints through a proxy layer.
What changed
noirdoc publicly launched its Claude Code privacy plugin, local pseudonymization engine, and hosted API proxy for broader LLM workflows.
Why it matters
This is the kind of privacy control many real-world AI coding deployments still lack, especially in consulting, finance, legal, and support workflows. If tools like noirdoc gain traction, pre-model redaction and reversible local mapping start looking like a standard security layer for agent stacks.
Evidence excerpt
noirdoc says its API proxy pseudonymizes requests to OpenAI, Anthropic, Azure, and OpenAI-compatible endpoints, while the open-source core and Claude Code plugin run locally under MIT licenses.