MCP Typesense

MCP server that gives AI tools search and retrieval over Typesense-indexed Payload content

Community 0 22/wk MIT v0.6.6

Installation

pnpm add @zetesis/mcp-typesense

About

MCP Typesense lets AI assistants and agents query content from a Typesense search index that backs a Payload site. It exposes search tools over the Model Context Protocol, so a connected model can run keyword, semantic and hybrid searches, read chunks and documents by ID or parent, and browse taxonomy trees and filter criteria.

Semantic matching runs inside Typesense through its auto-embed feature, so the package does not call an embedding service of its own. Results can be enriched with taxonomy data pulled from Payload, and the package includes optional tools that summarize documents, compare perspectives and extract claims using model sampling. Access control is pluggable, so the server can check who is calling before it returns content.

Package info

Package name
@zetesis/mcp-typesense
Latest version
0.6.6
Unpacked size
394 kB
License
MIT
Weekly downloads
22
Last publish
Jul 23, 2026

More from Zetesis-Labs

View profile

Tracks token usage, cost and sessions for LLM agent runs

Adds AI agent management, streaming chat endpoints and gateway-backed model access to Payload

Vector search adapter that stores embeddings in Postgres with pgvector

Keeps Payload collections synced to Typesense with keyword, semantic and hybrid search

Hierarchical taxonomy collection and relationship field for Payload collections

Builds Lexical rich text editors from feature and block lists, with typed block renderers

Agent skill that enforces generated, paired Payload database migrations and gates destructive changes

AI agent skill that guides building Payload blocks and their rendering dispatcher

Lets AI agents and scripts read, edit and sync Payload content from the command line

Cursor rules and a Payload override skill that keep AI agents on Payload conventions

Reusable Cursor coding-agent rules for Bun, commits, TypeScript and clean code

Command-line SEO checks and REST sync for Payload content, built for AI agents