Skip to content
MCP server

Trail

Issue tracker and shared project memory your coding agent reads and writes

Runs locallyStatus and history
Pricing
Freemium
Platforms
CLI, Web, macOS
Runs
Locally
Version
1.29.1
Listing updated
1 day ago

About Trail

Trail is a freemium MCP server for developer tools. It runs locally on your machine and is started by your AI client.

Trail gives a coding agent a real team board and a store of durable project knowledge, both scoped to the folder you have open. The agent reads open items, records what it did and skipped, and logs decisions as it works, so a status update does not depend on anyone remembering. Every change on a card keeps its previous value, the time, and whether a person or their agent made it.

Project memory holds things like a release command that wipes a database, or a rule that prices are integer cents. These entries are shared by everyone on the project, and a session reads them before acting instead of rediscovering them. Reminders carry dates the agent closes or reschedules itself, and attachments hang off items.

The package is a client that talks to your Trail account over REST. On login it can install a session-start hook for Claude Code, Gemini CLI or Antigravity that loads open items and memory. Clients can get read-only logins with progress views, a request box and a signed PDF of what shipped. The free plan covers three projects and three people.

Install

Run in a terminal

claude mcp add trail -- npx -y usetrail

Add to claude_desktop_config.json

{
  "mcpServers": {
    "trail": {
      "command": "npx",
      "args": [
        "-y",
        "usetrail"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "trail": {
      "command": "npx",
      "args": [
        "-y",
        "usetrail"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "trail": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "usetrail"
      ]
    }
  }
}

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

{
  "mcpServers": {
    "trail": {
      "command": "npx",
      "args": [
        "-y",
        "usetrail"
      ]
    }
  }
}

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

{
  "mcpServers": {
    "trail": {
      "command": "npx",
      "args": [
        "-y",
        "usetrail"
      ]
    }
  }
}

Configuration

  • TRAIL_API_URLOnly for a self-hosted Trail. Left unset, the connector points at the hosted service.
  • TRAIL_PROJECTOverrides the project name, which otherwise comes from the folder you have open.

Tools this server provides (18)

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

list_items
Lists tracker items.
get_item
Fetches one item.
add_item
Creates an item.
update_item
Updates an item.
move_item
Moves an item between statuses.
get_next
Picks the next item to work on.
delete_item
Deletes an item.
list_memory
Lists project memory entries.
get_memory
Reads one memory entry.
add_memory
Adds a memory entry.
update_memory
Edits a memory entry.
review_memory
Reviews memory entries.
add_reminder
Creates a reminder.
list_reminders
Lists reminders.
update_reminder
Updates a reminder.
delete_reminder
Deletes a reminder.
add_attachment
Attaches a file to an item.
get_attachment
Fetches an attachment.

What it does

Agent-written tracker
Items with type, status, priority, assignees and history.
Project memory
Shared gotchas and decisions read at the start of each session.
Folder-scoped projects
Uses the open folder, a .trail file or an environment variable.
Reminders
Dated items the agent closes or reschedules.
Session hook
Loads open items and memory into Claude Code, Gemini CLI or Antigravity.
Client view
Read-only progress login with a request box.

Good for

  • Record what an agent skipped or decided during a session
  • Warn future sessions about a dangerous release command
  • Give a client read-only progress without status emails
  • Track dated follow-ups an agent must not forget
  • Keep two repositories' work separate

Pricing

Freemium

Free plan: 3 projects, 3 people; paid plans count projects, not people

Prices change. Check usetrail.dev before you buy.

Trail FAQ

How do I connect it?

Run npx -y usetrail@latest login, then register it with your agent, for example claude mcp add trail -s user -- npx -y usetrail@latest.

What does the free plan include?

Three projects and three people, per the README.

How does it pick a project?

From the open folder name, a .trail file, or the TRAIL_PROJECT variable.

Can I self-host?

The TRAIL_API_URL variable points the connector at a self-hosted Trail.

Is Trail free?

Trail has a free plan, with paid plans for more usage or features.

How do I add Trail to Claude Code?

Run this command in your terminal: claude mcp add trail -- npx -y usetrail

What are the best alternatives to Trail?

The closest alternatives to Trail are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Trail

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. 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
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. 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
  6. 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