Skip to content
MCP server

ResumeState

Session continuity MCP server that records handoffs so an AI resumes where work stopped

Runs locallyStatus and history
Pricing
Free trial
Platforms
macOS, Windows, Linux
Runs
Locally
Version
1.0.0
Listing updated
2 days ago

About ResumeState

ResumeState is an MCP server for knowledge and memory. It runs locally on your machine and is started by your AI client.

The core concept is the lane, a named line of work. After a session the assistant writes a structured handoff to a lane (what was tried, decided, done and failed, plus current state and a pointer), and the next session, on any model or vendor, fetches it with resume instead of starting cold.

Sign-in uses the OAuth device flow: the assistant runs connect, shows a link and short code, and you approve it in your ResumeState account. The token is stored in a user-only file under your home directory, is narrow in scope and can be switched off under Connected AIs. Other tools list lanes, browse artifact cards before loading full content, and search a replay pool for routines that worked.

The client is MIT licensed and runs locally over stdio via uvx or a Claude Desktop bundle, but it talks to the ResumeState service, which requires a paid membership.

Install

Run in a terminal

claude mcp add resumestate-mcp -- npx -y https://github.com/Resume-State/resumestate-mcp/releases/download/v1.0.0/resumestate-mcp-1.0.0.mcpb

Add to claude_desktop_config.json

{
  "mcpServers": {
    "resumestate-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Resume-State/resumestate-mcp/releases/download/v1.0.0/resumestate-mcp-1.0.0.mcpb"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "resumestate-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Resume-State/resumestate-mcp/releases/download/v1.0.0/resumestate-mcp-1.0.0.mcpb"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "resumestate-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Resume-State/resumestate-mcp/releases/download/v1.0.0/resumestate-mcp-1.0.0.mcpb"
      ]
    }
  }
}

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

{
  "mcpServers": {
    "resumestate-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Resume-State/resumestate-mcp/releases/download/v1.0.0/resumestate-mcp-1.0.0.mcpb"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "resumestate-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Resume-State/resumestate-mcp/releases/download/v1.0.0/resumestate-mcp-1.0.0.mcpb"
      ]
    }
  }
}

Tools this server provides (11)

Taken from the project documentation; the running server may differ.

connect
Links the assistant to your account with an approval code.
resume
Fetches a lane's latest handoff or the account-wide packet.
list_lanes
Lists lanes with last handoff and freshness.
handoff
Records a structured handoff under a lane.
list_artifacts
Shows artifact cards.
get_artifact
Gets full artifact content.
replay_search
Finds routines that worked in the replay pool.
replay_get
Fetches a replay routine.
log_session
Logs session bookkeeping.
ingest
Ingests a transcript.
status
Shows connection status.

What it does

Named lanes
Organises continuity by line of work so each resumes separately.
Structured handoffs
Records tried, decided, done, failed, state and pointer.
Cross-model resume
Lets a different model or vendor pick up the latest handoff.
Device-flow sign-in
Approve the assistant in your account; no API key in config.
Artifact cards
Shows summaries first and full content only on request.
Replay search
Finds routines that worked in earlier sessions.

Good for

  • Continue a project in a fresh chat without re-explaining it
  • Switch models mid-project and keep the context
  • Hand work between teammates on the Team plan
  • Look up a routine that worked before

Pricing

Free trial

Membership required: Solo $9.99 CAD/mo, Team $19.99 CAD/mo, Enterprise $99 CAD/mo; 7-day trial

Prices change. Check resumestate.com before you buy.

ResumeState FAQ

What does it cost?

Plans are Solo at $9.99 CAD a month, Team at $19.99 CAD and Enterprise at $99 CAD, each with a 7-day trial.

Do I put an API key in my config?

No. The connect tool uses OAuth device flow.

What does it need to run?

Python 3.10 or later and uv.

Where is the token stored?

In a user-only file under ~/.resumestate.

Is ResumeState free?

ResumeState is a paid product with a free trial.

How do I add ResumeState to Claude Code?

Run this command in your terminal: claude mcp add resumestate-mcp -- npx -y https://github.com/Resume-State/resumestate-mcp/releases/download/v1.0.0/resumestate-mcp-1.0.0.mcpb

What are the best alternatives to ResumeState?

The closest alternatives to ResumeState are Codebase Memory, XERJ, Ha MCP, Claude Code Ultimate Guide and Nautilus Compass Memory.

Alternatives to ResumeState

  1. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  2. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  3. Ha MCP

    MCP server

    MCP server for managing and automating Home Assistant from AI assistants

    Runs locally
    • 8 tools
    • Open source
    • 7.9K GitHub stars
    • Productivity
  4. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  5. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  6. Concord MCP

    MCP server

    Local-first MCP server for messaging and shared work-state between coding agents

    Runs locally
    • Freemium
    • 1.3K GitHub stars
    • Developer tools