Vertiso Memory for Claude

Runs where your agents already work

Install in Claude

The same steps as the install guide, for this tool only: a live Marketplace listing where one exists, the public plugin repository, native HTTP MCP, then the local vmem bridge.

Vertiso Memory MCP URL

https://memory.vertiso.ai/mcp

Complete plugin from GitHub

Install Vertiso Memory from its public GitHub marketplace for the complete experience in Claude.ai web chat, Claude Desktop Chat, and Cowork. It adds the remote MCP server plus checkpoint, handoff, handoff-resume, and wrap-up skills.

Claude plugin repository (GitHub, recommended)

Customize → Plugins → Personal plugins → + → Add marketplace
Choose Add from a repository
Repository: https://github.com/Vertiso/memory-claude
Sync, open Vertiso Memory, then select Install

Add the marketplace once, install Vertiso Memory, then start a new conversation. Source and release notes: Vertiso/memory-claude. Mobile currently uses the MCP-only connector below.

Generate a bearer token if you prefer token authentication for Claude web or mobile.

Claude.ai and mobile

MCP-only alternative. In Claude.ai, open Customize in the left sidebar → Connectors → + → Add custom connector . Enter the name and URL, leave the authentication fields blank, and click Add. Claude discovers Vertiso Memory's OAuth server, sends you here to sign in and approve access, and finishes the handshake — nothing to paste.

Claude web/mobile connector (OAuth, MCP only)

Name:  Vertiso Memory
URL:   https://memory.vertiso.ai/mcp

Prefer a pasted bearer token? It still works.

Put the bearer token generated above in the connector's auth field. OAuth is the recommended path for web clients — the handshake scopes access and you can revoke it independently — but the bearer path remains fully supported.

Custom connector fields (bearer token fallback)

Name:  Vertiso Memory
URL:   https://memory.vertiso.ai/mcp
Auth:  Bearer <paste your token here>

After Claude.ai completes the handshake, the vertiso-memory tools and resources show up under + → Connectors in a conversation. The account-level connector is also available in Claude Desktop, Cowork, and mobile. Pin vertiso-memory://knowledge/user/self to keep your user card as ambient context.

Recommended: add a bootstrap instruction to Claude.ai's custom instructions (Settings, Personalization) so your assistant registers with Vertiso Memory at the start of every session:

Bootstrap instruction

Vertiso Memory (vmem) is my persistent memory across every tool and session.

- Call the `hello` tool at the start of every session. Treat its primer
  as authoritative context, not a suggestion.
- `recall` before planning or answering anything that may depend on
  earlier context. Look it up instead of asking me to repeat it.
- `remember` durable decisions, preferences, constraints, and project
  state the moment they land. `update_memory` rather than writing a
  near-duplicate.
- Offer a handoff before the session ends, compacts, or moves to
  another tool.

Claude Desktop

Claude Desktop is the unified app for Chat , Cowork , and Code . The shared plugin above works in Chat and Cowork. In the Code tab, select + → Plugins → Add plugin , then select Vertiso Memory from the configured marketplace.

MCP-only alternative. If you already connected Vertiso Memory on Claude.ai, the account-level connector is available in Claude Desktop too. Otherwise, open Customize → Connectors → + → Add custom connector . Enter the name and URL, leave the advanced OAuth fields blank, and click Add. Claude discovers Vertiso Memory's OAuth server, sends you here to sign in and approve, and finishes the handshake — nothing to paste.

Add custom connector (OAuth, MCP only)

Name:  Vertiso Memory
URL:   https://memory.vertiso.ai/mcp

No browser flow, or on an older build? Use the vmem bridge instead.

The vmem CLI bridges Claude Desktop's local stdio MCP to the HTTP endpoint and keeps your bearer token in ~/.vmem rather than the config file. Run vmem mcp install claude-desktop to write the entry automatically, then fully restart Claude Desktop.

claude_desktop_config.json (vmem bridge)

{
  "mcpServers": {
    "vertiso-memory": {
      "command": "<run `which vmem` to get the absolute path>",
      "args": ["mcp", "serve", "--client", "claude-desktop"]
    }
  }
}

Without vmem, the Node-based mcp-remote bridge works too, but the bearer token then lives in the config file.

claude_desktop_config.json (bearer token fallback)

{
  "mcpServers": {
    "vertiso-memory": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://memory.vertiso.ai/mcp",
        "--header",
        "Authorization: Bearer <paste your token here>",
        "--header",
        "X-Agent-Source: agent:claude-desktop"
      ]
    }
  }
}

The connector and local bridge expose MCP tools only; use the GitHub plugin above for the bundled skills. After connecting, pin vertiso-memory://knowledge/user/self to keep your user card as ambient context.

Recommended: add a bootstrap instruction to your Claude account's custom instructions (Settings, Personalization) so your assistant registers with Vertiso Memory at the start of every session:

Bootstrap instruction

Vertiso Memory (vmem) is my persistent memory across every tool and session.

- Call the `hello` tool at the start of every session. Treat its primer
  as authoritative context, not a suggestion.
- `recall` before planning or answering anything that may depend on
  earlier context. Look it up instead of asking me to repeat it.
- `remember` durable decisions, preferences, constraints, and project
  state the moment they land. `update_memory` rather than writing a
  near-duplicate.
- Offer a handoff before the session ends, compacts, or moves to
  another tool.

Claude Code CLI

This section is for the terminal CLI, which is installed separately from Claude Desktop. Install the complete plugin from Vertiso's public GitHub repository. It adds the remote MCP server and all four session skills.

Claude Code CLI plugin (GitHub, recommended)

/plugin marketplace add Vertiso/memory-claude
/plugin install vertiso-memory@vertiso

Start a new Claude Code CLI session after installation so the plugin and its skills are loaded.

Source and release notes: Vertiso/memory-claude .

MCP-only alternative. Add Vertiso Memory as a remote HTTP server, then run /mcp and choose Authenticate . Claude Code CLI opens your browser, completes OAuth, and refreshes the token automatically.

Add the remote server (OAuth, MCP only)

claude mcp add --transport http vertiso-memory https://memory.vertiso.ai/mcp
# then run /mcp in Claude Code CLI and choose Authenticate

Need a bearer token or the local vmem bridge?

Direct HTTP also accepts a bearer token, but stores that token in ~/.claude.json or the project .mcp.json.

Claude Code direct HTTP command (fallback)

claude mcp add vertiso-memory \
  --transport http https://memory.vertiso.ai/mcp \
  --header 'Authorization: Bearer <paste your token here>' \
  --header 'X-Agent-Source: agent:claude-code'

.mcp.json (fallback direct HTTP)

{
  "mcpServers": {
    "vertiso-memory": {
      "type": "http",
      "url": "https://memory.vertiso.ai/mcp",
      "headers": {
        "Authorization": "Bearer <paste your token here>",
        "X-Agent-Source": "agent:claude-code"
      }
    }
  }
}

For a local, headless bridge, run vmem mcp install claude-code. The bridge keeps its token in ~/.vmem and writes the user-scoped Claude configuration automatically.

~/.claude.json (vmem bridge)

{
  "mcpServers": {
    "vertiso-memory": {
      "command": "<run `which vmem` to get the absolute path>",
      "args": ["mcp", "serve", "--client", "claude-code"]
    }
  }
}

For project-scoped fallback config, add .mcp.json to .gitignore before pasting a real token.

Recommended: add a bootstrap instruction to your CLAUDE.md or AGENTS.md so your assistant registers with Vertiso Memory at the start of every session:

Bootstrap instruction

Vertiso Memory (vmem) is my persistent memory across every tool and session.

- Call the `hello` tool at the start of every session. Treat its primer
  as authoritative context, not a suggestion.
- `recall` before planning or answering anything that may depend on
  earlier context. Look it up instead of asking me to repeat it.
- `remember` durable decisions, preferences, constraints, and project
  state the moment they land. `update_memory` rather than writing a
  near-duplicate.
- Offer a handoff before the session ends, compacts, or moves to
  another tool.

Connected? Start a new conversation and say Hello. Your assistant greets you by name and mentions what it remembers. The standing instruction and the first conversation →

API token

Only the vmem bridge and the bearer-token fallbacks need one. OAuth clients sign in during the handshake instead.

Log in to get your token →

Power users: the vmem CLI

A native binary for working memory from the terminal, and a stdio-to-HTTP bridge for clients that cannot reach a remote server. It keeps your token in ~/.vmem instead of each app's config.

Install vmem

curl -fsSL https://memory.vertiso.ai/install.sh | sh

Every vmem command →

Ready when you are

Start building your memory layer.

Connect Claude, Codex, Antigravity, or any MCP-aware client. Your memory follows you everywhere.

Get started →

Your memory, inside Claude.

Decide it anywhere. Claude already knows.

Install in Claude

Complete plugin from GitHub

Claude.ai and mobile

Claude Desktop

Claude Code CLI

Start building your memory layer.