MCP Server

Active

MIT

Perenna MCP Server

io.github.scarletkc/perenna

Git-backed permanent memory for AI agents, shared across Claude, Cursor, ChatGPT, and other MCP clients.

What is the Perenna MCP server?

Perenna is a lightweight MCP server that provides durable, Git-backed memory for AI agents. It stores human-readable Markdown in an independent Git repository, allowing Claude, Cursor, ChatGPT, and other clients to share memories without vendor lock-in or conversation silos. Memories persist across agents, machines, and sessions with local Vexor-powered retrieval and optional remote synchronization.

Perenna solves the problem of fragmented AI agent memory by providing a shared, self-hosted memory system backed by Git. Instead of memories locked in separate vendor silos, Perenna lets you maintain one durable memory store accessible from any MCP client. Memories are stored as ordinary Markdown you can inspect, edit, version, and back up yourself, with optional remote Git synchronization for multi-machine access.

How to install Perenna

Copy-paste configuration for popular MCP clients.

transport: stdio

Config generated by PluginBench — verify against the source before use.

Environment / auth

PERENNA_GIT_REMOTE
Git remote name used for optional synchronization (for example, origin); leave unset for local-only operation
VEXOR_API_KEY
secret
Embedding provider API key (optional; overrides a stored key; providers can also be configured via `vexor init`)
VEXOR_REMOTE_RERANK_API_KEY
secret
Remote reranker API key (optional; overrides a stored key; only needed when rerank is set to remote)
VEXOR_CONFIG_JSON
Non-secret Vexor config as a JSON object, e.g. {"provider": "gemini", "rerank": "bm25"} (optional; merged over ~/.vexor/config.json; credential fields are rejected, use the dedicated variables)

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "perenna": {
      "command": "uvx",
      "args": [
        "perenna",
        "mcp",
        "--home"
      ],
      "env": {
        "PERENNA_GIT_REMOTE": "<YOUR_PERENNA_GIT_REMOTE>",
        "VEXOR_API_KEY": "<YOUR_VEXOR_API_KEY>",
        "VEXOR_REMOTE_RERANK_API_KEY": "<YOUR_VEXOR_REMOTE_RERANK_API_KEY>",
        "VEXOR_CONFIG_JSON": "<YOUR_VEXOR_CONFIG_JSON>"
      }
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

read_memory — Retrieve memories from the permanent memory store using Vexor-powered semantic search.
write_memory — Store new memories or update existing ones in the Git-backed memory repository.
delete_memory — Remove memories from the permanent memory store.

Use cases

Share persistent memories across multiple AI agents (Claude, Cursor, ChatGPT) without vendor lock-in
Maintain a searchable knowledge base of project context, decisions, and learnings that survives agent switches
Self-host a private memory service for multiple machines or team members via Git synchronization
Build durable agent workflows that accumulate knowledge over time without relying on conversation history
Version control and audit all agent memories as ordinary Git-tracked Markdown files

Perenna MCP server FAQ

What is Perenna?

Perenna is a Git-backed permanent memory system for AI agents. It lets Claude, Cursor, ChatGPT, and other MCP clients share durable memories stored as Markdown in a Git repository, rather than keeping memories siloed in each agent's proprietary system.

Is Perenna free?

Yes, Perenna is open-source under the MIT license and free to use. You only need Python 3.12+, Git, and uv (a Python package manager), plus a Vexor embedding provider (which can be local or remote).

How do I install Perenna in Cursor or Claude?

Run `uv tool install perenna`, configure a Vexor embedding provider with `uvx vexor init`, then configure your MCP client to start `perenna mcp`. Detailed instructions are available in the agent-installation guide in the repository.

Does Perenna require authentication or a cloud account?

No. Perenna is designed to work locally without required accounts. It supports local stdio, loopback-only HTTP, and optional OAuth-protected remote HTTP for self-hosted setups, but no vendor account is required.

Can I sync memories across multiple machines?

Yes. Perenna supports optional Git remote synchronization via `perenna sync setup <repository-url>`, allowing you to import, publish, and fast-forward memory history through a private Git repository.

What embedding provider does Perenna use?

Perenna uses Vexor for semantic search and retrieval. You can configure a remote provider (OpenAI, Anthropic, etc.) or use local embeddings by installing the `perenna[local]` extra.

README (reference)

Source of truth, from the repository.

Perenna

</div>

A lightweight, Git-backed permanent memory for AI agents. Claude Code, Codex, ChatGPT, Cursor, and other MCP clients can share durable memories without sharing a vendor account or conversation history.

Separate MCP tools for reading, writing, and deleting memories
Local stdio, loopback-only Streamable HTTP, and OAuth-protected remote HTTP
Human-readable Markdown stored in an independent Git repository
Local Vexor retrieval index that can always be rebuilt from Git
Cross-process locking for multiple local agent processes

Why Perenna?

Your memory should follow you, not the agent you happen to be using.

Claude Code, Codex, Cursor, and ChatGPT keep memory in separate silos. Switch agents and your memory disappears. Switch machines and local memory stays behind.

Perenna gives them one shared, Git-backed memory. Local agents and ChatGPT can connect to the same self-hosted Perenna service, while every durable memory stays ordinary Markdown you can inspect, edit, version, and back up yourself.

Perenna stays focused on permanent memory: no required account, proprietary memory cloud, or automatic conversation extraction.

Quickstart

Install with your AI agent

Paste this into Claude Code, Codex, ChatGPT Desktop, Cursor, or another coding agent with terminal and local MCP configuration access:

Open the following URL, read the complete instructions, and follow them to
install and connect Perenna:
https://raw.githubusercontent.com/scarletkc/Perenna/main/docs/guides/agent-installation.md

Install a published release

Perenna requires Python 3.12+, Git, and uv.

Install Perenna

uv tool install perenna

Configure retrieval

Perenna needs a working Vexor embedding provider. For interactive provider selection and configuration, run:

uvx vexor init

Perenna automatically reuses ~/.vexor/config.json. When using process-level configuration, make sure the MCP server receives VEXOR_CONFIG_JSON plus VEXOR_API_KEY or the selected provider's key from its host environment. Remote providers receive memory text and search queries.

If you choose local embeddings, also install Perenna's local extra:

uv tool install "perenna[local]"

Vexor provider configuration covers remote and local setup. From the environment that starts the MCP client, verify the selected provider with:

uvx vexor doctor

Connect a client

For the standalone setup, configure the MCP client to start:

perenna mcp

Follow the Client setup guide for the exact client command or configuration file.

Codex and Claude Code can also install the optional memory behavior Skill:

perenna skill install --agent codex
# or
perenna skill install --agent claude-code

Repeat --agent in one command when both clients should receive the skill. The configuration reference documents user and project scope, destinations, and replacement safeguards.

Codex and Claude Code can instead install the combined Skill and MCP connection from Perenna's repository Marketplace. Follow the Plugin setup guide and choose one setup path per client.

Perenna creates its local data under ~/.perenna/ unless another home is configured.

Add optional Git synchronization

To import, publish, or fast-forward compatible history through a private Git repository, run:

perenna sync setup <repository-url>

Successful setup saves the selected remote in the Perenna home. Use perenna sync disable to return to saved local-only mode without removing the Git remote.

The configuration reference owns remote selection, credentials, conflict handling, and recovery details.

Install from source

git clone https://github.com/scarletkc/Perenna.git
cd Perenna
uv tool install .

Source contributors should use the locked environment in the development guide.

Documentation

Start with the documentation index, then follow the path for your task:

License

MIT

Related MCP servers

Vexor

Active

Semantic search engine for files and code with configurable embeddings and reranking.

237

Python

MIT

View repository →

Scavio

Active

Structured web data from 31 platforms: Google, YouTube, Amazon, Walmart, Reddit, TikTok, LinkedIn

TypeScript

MIT

View repository →

Quipu

Active

Local knowledge graph memory for agents. SHACL rules refuse bad facts; SPARQL and time travel.

Rust

MIT

View repository →

io.github.sceneview/mcp

Active

3D & AR SDK for Android, iOS, Web, Desktop, Flutter, React Native — build with Compose, SwiftUI, or JavaScript.

1.3k

Kotlin

Apache-2.0

View repository →

Lodging Decision

Active

Agent-facing lodging decision tool. Non-sensitive categories only. Alpha, unauthenticated.

TypeScript

View repository →

io.github.schemacrawler/schemacrawler-ai

Active

Natural language database schema exploration and SQL assistance via AI agents

Java

View repository →