ax128ax128Communityยท1 Agent Skills Included

AegisGate

Self-hosted security gateway filtering LLM API traffic

Filters LLM API requests and responses through a self-hosted security gateway with PII redaction, prompt injection detection, and dangerous output blocking. Eliminates scattered per-app security code by centralizing policies, token routing, and audit logging in one proxy. Deploys via Docker Compose with a web console for managing tokens, rules, and real-time request statistics.
npx skills add ax128/AegisGate --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install AegisGate?โ–ผ

Run `npx skills add ax128/AegisGate --all -g -y` in your terminal to install this skill globally. Then start the gateway with Docker Compose as described in the setup steps.

What does AegisGate protect against?โ–ผ

It blocks prompt injection attacks, redacts over 50 categories of PII and secrets like API keys and credit cards, and sanitizes dangerous LLM outputs such as malicious shell commands before they reach your apps.

Does AegisGate work with OpenAI and Claude?โ–ผ

Yes. It provides drop-in OpenAI-compatible routes and an Anthropic Messages endpoint, and can even convert between the two protocols so Claude Code can talk to OpenAI-compatible upstreams.

Can I use AegisGate with Cursor or Claude Code?โ–ผ

Yes. Point your coding agent's baseUrl at the gateway and all its LLM traffic is filtered automatically, with MCP support for Cursor, Claude Code, Codex, and Windsurf.

Does AegisGate need an external detection service?โ–ผ

No. Core protections like regex-based injection detection, redaction, and response sanitization run locally, and the optional semantic review service is only used for gray-zone cases.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’