tiretrack-mcp 0.2.5.dev20260611202536

MCP server for TireTrack project management — manage projects, issues, sprints, and teams via AI

tiretrack-mcp

MCP server for TireTrack project management. Manage projects, issues, sprints, and teams from any AI assistant that supports the Model Context Protocol.

The TireTrack Settings → MCP Server Setup page renders the exact config block for your environment and chosen client (Claude Code / Codex CLI / Gemini CLI). Use that as the canonical source — this README is a reference for offline setup.

Tools

Category	Tools
Projects	list_projects, get_project, create_project
Members	list_project_members, add_project_member
Issues	list_issues, get_issue, get_issue_by_key, create_issue, update_issue
Comments	list_comments, add_comment
Sprints	list_sprints, get_sprint_metrics
Search	search_issues
Users	list_users, get_current_user

Quick Start

Using uvx (recommended)

uvx tiretrack-mcp

Using pip

pip install tiretrack-mcp
tiretrack-mcp

Configuration

Set these environment variables:

Variable	Required	Description
TIRETRACK_API_URL	Yes	API base URL — https://api.tiretrack.mi.tirerack.com (prod) or https://api-dev.tiretrack.mi.tirerack.com (dev)
TIRETRACK_API_TOKEN	Yes	Personal access token minted at Settings → Security → Personal Access Tokens

Tokens look like tt_pat_…, expire after 90 days by default, and can be revoked at any time from the same page.

The examples below use ${TIRETRACK_API_TOKEN} so the value resolves from your shell environment. If you'd rather not export it, paste the token string directly into the TIRETRACK_API_TOKEN field instead.

Claude Code / Claude Desktop

Stable (production):

{
  "mcpServers": {
    "tiretrack": {
      "command": "uvx",
      "args": ["--refresh", "tiretrack-mcp"],
      "env": {
        "TIRETRACK_API_URL": "https://api.tiretrack.mi.tirerack.com",
        "TIRETRACK_API_TOKEN": "${TIRETRACK_API_TOKEN}"
      }
    }
  }
}

Dev (pre-release):

{
  "mcpServers": {
    "tiretrack": {
      "command": "uvx",
      "args": ["--refresh", "--prerelease", "allow", "tiretrack-mcp"],
      "env": {
        "TIRETRACK_API_URL": "https://api-dev.tiretrack.mi.tirerack.com",
        "TIRETRACK_API_TOKEN": "${TIRETRACK_API_TOKEN}"
      }
    }
  }
}

--refresh checks PyPI for newer versions on each invocation. --prerelease allow installs .dev pre-release versions.

OpenAI Codex CLI

Stable (production):

# .codex/config.toml
[mcp_servers.tiretrack]
command = "uvx"
args = ["--refresh", "tiretrack-mcp"]

[mcp_servers.tiretrack.env]
TIRETRACK_API_URL = "https://api.tiretrack.mi.tirerack.com"
TIRETRACK_API_TOKEN = "${TIRETRACK_API_TOKEN}"

Dev (pre-release):

# .codex/config.toml
[mcp_servers.tiretrack]
command = "uvx"
args = ["--refresh", "--prerelease", "allow", "tiretrack-mcp"]

[mcp_servers.tiretrack.env]
TIRETRACK_API_URL = "https://api-dev.tiretrack.mi.tirerack.com"
TIRETRACK_API_TOKEN = "${TIRETRACK_API_TOKEN}"

Gemini CLI

Stable (production):

{
  "mcpServers": {
    "tiretrack": {
      "command": "uvx",
      "args": ["--refresh", "tiretrack-mcp"],
      "env": {
        "TIRETRACK_API_URL": "https://api.tiretrack.mi.tirerack.com",
        "TIRETRACK_API_TOKEN": "${TIRETRACK_API_TOKEN}"
      }
    }
  }
}

Dev (pre-release):

{
  "mcpServers": {
    "tiretrack": {
      "command": "uvx",
      "args": ["--refresh", "--prerelease", "allow", "tiretrack-mcp"],
      "env": {
        "TIRETRACK_API_URL": "https://api-dev.tiretrack.mi.tirerack.com",
        "TIRETRACK_API_TOKEN": "${TIRETRACK_API_TOKEN}"
      }
    }
  }
}

Transports

Transport	Command	Use case
stdio (default)	tiretrack-mcp	Claude Code, Codex, Gemini CLI, Claude Desktop
Streamable HTTP	tiretrack-mcp --transport streamable-http --port 8080	ChatGPT, remote clients
SSE (legacy)	tiretrack-mcp --transport sse --port 8080	Older clients

Development

cd mcp-server
uv sync
uv run tiretrack-mcp --help