n24q02m/wet-mcp

v2.16.0 Security

This release includes 1 security fix for security teams reviewing exposed deployments.

Published 2mo MCP Search & Web

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

This release patches 1 known CVE

Topics

ai-agents ai-coding claude claude-code crawl4ai cursor

+8 more

docker mcp mcp-server model-context-protocol python searxng web-scraping web-search

Affected surfaces

deps

Summary

AI summary

Complete removal of LiteLLM dependency due to supply chain attack.

Full changelog

What's Changed

Multi-provider embedding & LLM

Embedding: Jina > Gemini > OpenAI > Cohere (auto-detect from API keys), local ONNX fallback
Reranking: Jina > Cohere (cloud), local ONNX fallback
LLM: Gemini (gemini-3-flash-preview) with OpenAI fallback (gpt-5.4-mini-2026-03-17)

LiteLLM removal

Complete removal of LiteLLM dependency (supply chain attack response)
Native provider SDKs: google-genai, openai, cohere, httpx (Jina)

Default models

Embed: jina-embeddings-v5-text-small > gemini-embedding-001 > text-embedding-3-large > embed-multilingual-v3.0
Rerank: jina-reranker-v3 > rerank-v4.0-pro

Breaking Changes

Removed LiteLLM dependency entirely.

Security Fixes

Removed LiteLLM dependency as a response to a supply chain attack.

View diff on GitHub

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.

Share this release

Share on X Share on Bluesky

Track n24q02m/wet-mcp

Get notified when new releases ship.

About n24q02m/wet-mcp

Web search (embedded SearXNG), content extraction, and library docs indexing with hybrid search (FTS5 + semantic). Built-in Qwen3 embedding, no API keys required.

All releases →