Skip to content

Nenya

LLM Frameworks

A lightweight, zero‑dependency Go API gateway that proxies AI model requests while adding secret redaction, context management, routing, and MCP tool integration.

Track releases GitHub
Go Latest v0.8.6 · 9d ago Security brief →

Features

  • Config‑driven provider registry with 23 built‑in adapters for OpenAI/Anthropic compatible APIs
  • Dynamic model discovery and multi‑provider fallback chains with latency‑aware routing
  • Pluggable security interceptor chain (regex secret redaction, entropy filtering, TF‑IDF scoring, summarization) and RBAC enforcement

Recent releases

View all 21 releases →
v0.8.6 Feature
Notable features
  • Thinking‑aware stall timeout extension for streams
Full changelog

Changelog

  • fe503f3212c802bb3370defa797a0df8accd560f feat(stream): add thinking-aware stall timeout extension
View release on GitHub
Config change
v0.8.5 Security relevant
Auth

Redacted credentials + headers

Open
No immediate action
v0.8.4 Bugfix

GLM-4.7 SSE truncation fix

Open
No immediate action
v0.8.3 Bug fix

Proxy refusal handling

Open
Review required
v0.8.2 New feature
Auth RCE / SSRF

Thinking config + nested tokens + provider check

Open

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

About

Stars
25
Forks
0
Languages
Go Shell Dockerfile
View on GitHub

Similar tools

Sediman Browse
Deja
Noi
Klimkit
Rudo

Beta — feedback welcome: [email protected]