← All tools
LLM frameworks · pgEdge

pgEdge AI toolkit

pgEdge's open-source RAG server and Go LLM library for building AI apps on PostgreSQL and pgvector.

1Advisories
0critical
1high
0medium
2026-11-10Latest
Timeline · advisories per quarter
12026 Q4
Named in sources

Reported for pgEdge AI toolkit

01. highData leakagepgEdge RAG Server before 2.0.0 and Go LLM Library 0.1.x: unauthenticated DoS, provider API key fragments in errors, results leaking between requests (CVE-2026-125325 to 125328)pgEdge RAG Server (retrieval-augmented generation over PostgreSQL/pgvector) 1.0.0 and earlier exposes an unauthenticated query endpoint with three flaws fixed in 2.0.0: each request reads the whole corpus and rebuilds a BM25 index (DoS), failed LLM calls return provider error text that quotes part of the operator's API key, and a shared keyword index can return one caller's filtered documents to another. pgEdge Go LLM Library 0.1.0-0.1.1 has the same key-fragment leak through its unauthenticated /v1/health endpoint, fixed in 0.2.0. pgEdge AI toolkit 2026-11-10
The weekly email

One email a week. Only what changed.

Every new security hole in AI coding tools, MCP servers, app builders and the stacks they generate (Next.js, Supabase, Node, PHP), each with how to check your app and the fix. No roundups of things you already read.

Free, one email a week. Read the latest issue first if you like. One-click unsubscribe in every email. We never sell or share addresses.