Shield · LLM Security Gateway & Platform
Redacts secrets and PII from your prompts before they reach any LLM, then restores them
Sits in front of every LLM tool your team uses — Claude Code, Codex, Cursor, DeepSeek, Ollama, plus desktop apps like Claude Desktop, VS Code Copilot and Zed — and swaps secrets and PII matched by its filter packs for placeholders before they leave your machine, then rehydrates them back for the developer. Runs as one local binary — gateway, CLI, dashboard and audit log — with no hosted service and no data pipeline, and catches leaks twice: a plugin at the agent-loop boundary and the gateway scrubbing the wire. Sold either as a source-delivered platform engagement (Foundation, or Compliance with hash-chained audit trails and SOC 2 / HIPAA evidence export) or a self-serve subscription app running the same binary.
What holds up
- +Runs as one local binary — no hosted service, no data leaving your infrastructure
- +Catches leaks twice: plugin at the agent loop, gateway on the wire — one layer's miss is the other's catch
- +Audit trail is hash-chained and tamper-evident, with one-command SOC 2 / HIPAA evidence export
Mind the limits
- −Pricing for the platform tiers is scoped per engagement — no listed price, needs a call
- −Semantic detection beyond pattern matching is a separate, higher Custom tier
Featured here? Take the badge
Put it on your site — it links back to this review. Free for every listed product, always.
<a href="https://stillworks.dev/products/p/shield-llm-security-gateway-platform/"><img src="https://stillworks.dev/badge/shield-llm-security-gateway-platform.svg" alt="Picked by StillWorks" width="250" height="54"></a>