SERVER

Cathedral — Persistent Memory for AI Agents

Rank #36,219

io.github.AILIFE1/cathedral-mcp

Persistent memory and drift detection for AI agents across session restarts.

2 versions

11 GitHub stars

by ailife1 (2 servers)View repository ↗

First listed

Apr 7, 2026

Last publish date

Apr 12, 2026

PURPOSE

What you can do with Cathedral — Persistent Memory for AI Agents

Cathedral — Persistent Memory for AI Agents is a Model Context Protocol (MCP) server published by ailife1. It ranks #36,219 of 167,945 active servers on MCP Toplist, and its repository has 11 GitHub stars. Cathedral — Persistent Memory for AI Agents is listed across 3 registries — Official MCP, PulseMCP and Glama, with 2 tracked versions. It was first listed on Apr 7, 2026 and most recently updated on Apr 12, 2026.

Not verifiedMCP Toplist has not yet written a checked profile for this server, so the paragraph above is assembled from registry data only. The repository README is the authority on what it does.

CAPABILITIES AND LIMITATIONS

What Cathedral — Persistent Memory for AI Agents can and cannot do

Not verifiedMCP Toplist has not catalogued this server's tools or limitations. The repository README lists them.

REQUIREMENTS

What you need before using Cathedral — Persistent Memory for AI Agents

DeclaredNo credentialsNo credentials declared

The registry listing declares no required environment variables or authentication. This is declared metadata, not a verified install — the repository README is the authority.

MCP Toplist probed the public endpoint on Oct 28, 2026 and could not complete the MCP handshake (jsonrpc error). Requirements remain unverified.

Declared by Glama, last synced Nov 3, 2026. Not independently verified by MCP Toplist.

Declared in the registry install config

CATHEDRAL_API_KEY · required
CATHEDRAL_BASE_URL · optional
CATHEDRAL_SANITISE · optional

Hosted endpoint declared as public.

INSTALLATION AND FIRST TASK

Connect Cathedral — Persistent Memory for AI Agents and confirm it works

Remote · HTTP

claude mcp add --transport http cathedral-mcp https://cathedral-ai.com/mcp

Python package (PyPI)

claude mcp add cathedral-mcp -e "CATHEDRAL_API_KEY=<your CATHEDRAL_API_KEY>" -e "CATHEDRAL_BASE_URL=<your CATHEDRAL_BASE_URL>" -e "CATHEDRAL_SANITISE=<your CATHEDRAL_SANITISE>" -- uvx cathedral-mcp

Replace each placeholder with your own value before saving.

Remote · HTTP

[mcp_servers.cathedral-mcp]
url = "https://cathedral-ai.com/mcp"

Add to ~/.codex/config.toml.

Python package (PyPI)

[mcp_servers.cathedral-mcp]
command = "uvx"
args = ["cathedral-mcp"]

[mcp_servers.cathedral-mcp.env]
CATHEDRAL_API_KEY = "<your CATHEDRAL_API_KEY>"
CATHEDRAL_BASE_URL = "<your CATHEDRAL_BASE_URL>"
CATHEDRAL_SANITISE = "<your CATHEDRAL_SANITISE>"

Add to ~/.codex/config.toml. Replace each placeholder with your own value before saving.

Remote · HTTP

https://cathedral-ai.com/mcp

Settings → Connectors → “Add custom connector”, then paste this URL.

Python package (PyPI)

{
  "mcpServers": {
    "cathedral-mcp": {
      "command": "uvx",
      "args": [
        "cathedral-mcp"
      ],
      "env": {
        "CATHEDRAL_API_KEY": "<your CATHEDRAL_API_KEY>",
        "CATHEDRAL_BASE_URL": "<your CATHEDRAL_BASE_URL>",
        "CATHEDRAL_SANITISE": "<your CATHEDRAL_SANITISE>"
      }
    }
  }
}

Add to claude_desktop_config.json (Settings → Developer → Edit Config). Replace each placeholder with your own value before saving.

Remote · HTTP

{
  "mcpServers": {
    "cathedral-mcp": {
      "url": "https://cathedral-ai.com/mcp"
    }
  }
}

Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (per project).

Python package (PyPI)

{
  "mcpServers": {
    "cathedral-mcp": {
      "command": "uvx",
      "args": [
        "cathedral-mcp"
      ],
      "env": {
        "CATHEDRAL_API_KEY": "<your CATHEDRAL_API_KEY>",
        "CATHEDRAL_BASE_URL": "<your CATHEDRAL_BASE_URL>",
        "CATHEDRAL_SANITISE": "<your CATHEDRAL_SANITISE>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (per project). Replace each placeholder with your own value before saving.

Remote · HTTP

{
  "servers": {
    "cathedral-mcp": {
      "type": "http",
      "url": "https://cathedral-ai.com/mcp"
    }
  }
}

Save as .vscode/mcp.json in your workspace.

Python package (PyPI)

{
  "servers": {
    "cathedral-mcp": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "cathedral-mcp"
      ],
      "env": {
        "CATHEDRAL_API_KEY": "<your CATHEDRAL_API_KEY>",
        "CATHEDRAL_BASE_URL": "<your CATHEDRAL_BASE_URL>",
        "CATHEDRAL_SANITISE": "<your CATHEDRAL_SANITISE>"
      }
    }
  }
}

Save as .vscode/mcp.json in your workspace. Replace each placeholder with your own value before saving.

Remote · HTTP

{
  "mcpServers": {
    "cathedral-mcp": {
      "serverUrl": "https://cathedral-ai.com/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json.

Python package (PyPI)

{
  "mcpServers": {
    "cathedral-mcp": {
      "command": "uvx",
      "args": [
        "cathedral-mcp"
      ],
      "env": {
        "CATHEDRAL_API_KEY": "<your CATHEDRAL_API_KEY>",
        "CATHEDRAL_BASE_URL": "<your CATHEDRAL_BASE_URL>",
        "CATHEDRAL_SANITISE": "<your CATHEDRAL_SANITISE>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json. Replace each placeholder with your own value before saving.

Remote · HTTP

{
  "mcpServers": {
    "cathedral-mcp": {
      "url": "https://cathedral-ai.com/mcp"
    }
  }
}

Python package (PyPI)

{
  "mcpServers": {
    "cathedral-mcp": {
      "command": "uvx",
      "args": [
        "cathedral-mcp"
      ],
      "env": {
        "CATHEDRAL_API_KEY": "<your CATHEDRAL_API_KEY>",
        "CATHEDRAL_BASE_URL": "<your CATHEDRAL_BASE_URL>",
        "CATHEDRAL_SANITISE": "<your CATHEDRAL_SANITISE>"
      }
    }
  }
}

Replace each placeholder with your own value before saving.

First successful task

After adding the config, ask your client to list this server's tools. A tool list confirms the connection; an empty list or an error means the server did not start or needs credentials.

MAINTENANCE AND EVIDENCE

What changed, and when this was checked

First listed: Apr 7, 2026
Latest version published: Apr 12, 2026
Version history: 2 versions tracked (below)
Requirements evidence: MCP Toplist probed the public endpoint on Oct 28, 2026 and could not complete the MCP handshake (jsonrpc error). Requirements remain unverified.
Profile checked: Not yet — purpose, capabilities and requirements above come from registry data only
Ranking data as of: Nov 3, 2026

Registry sightings

Official MCP ↗listed since Apr 25, 2026 · still listed
PulseMCP ↗listed since Jun 4, 2026 · still listed
Glama ↗listed since Jun 4, 2026 · still listed

Do you maintain Cathedral — Persistent Memory for AI Agents? to correct or enrich this profile.

VERSIONS

Available versions (2)

Version	Published
1.1.0	Apr 12, 2026
1.0.1	Apr 7, 2026

STANDING

#36,219of 167,945 active serversTop 25%

Ranks ahead of 131,726 of 167,945 servers on MCP Toplist.

BADGE

Show your rank

Maintain this server? Add the live rank badge to your README — it updates automatically as the leaderboard changes.

Markdown

[![MCP Toplist](https://mcptoplist.com/badge/io.github.AILIFE1%2Fcathedral-mcp.svg)](https://mcptoplist.com/server/io.github.AILIFE1%2Fcathedral-mcp)

HTML

<a href="https://mcptoplist.com/server/io.github.AILIFE1%2Fcathedral-mcp"><img src="https://mcptoplist.com/badge/io.github.AILIFE1%2Fcathedral-mcp.svg" alt="MCP Toplist: Top 25% of 167,945" /></a>

Variants: append ?metric=score or ?metric=stars to the image URL.

FAQ

Frequently asked questions

Who maintains Cathedral — Persistent Memory for AI Agents?: Cathedral — Persistent Memory for AI Agents is maintained by ailife1, which publishes 2 MCP servers (2 total versions) tracked on MCP Toplist.
Is Cathedral — Persistent Memory for AI Agents listed on the Official MCP Registry?: Yes — Cathedral — Persistent Memory for AI Agents is listed on the Official MCP Registry, alongside PulseMCP and Glama.
How many versions does Cathedral — Persistent Memory for AI Agents have?: MCP Toplist tracks 2 versions of Cathedral — Persistent Memory for AI Agents, most recently published on Apr 12, 2026.
Where can I find the source code for Cathedral — Persistent Memory for AI Agents?: The source code for Cathedral — Persistent Memory for AI Agents is hosted at github.com/ailife1/cathedral.

Used in projects

Do you run this server?

Community projects list the MCP servers behind real agents and apps. Publish yours and this page will link to it.

EXPLORE

All MCP Servers Official MCP Servers GitHub MCP Servers Best MCP Servers Most Active MCP Servers New MCP Servers Best New MCP Servers