Skip to main content
Use tk with Zed to manage issues from a high-performance, Rust-native code editor with built-in AI and MCP support.

What is Zed?

Zed is a high-performance code editor built in Rust with native AI and real-time collaboration features. It supports MCP servers through its Agent Panel, giving AI assistants direct access to external tools. As a Rust-native editor, Zed is a natural fit for the Rust-built tk.

MCP Integration

Install the ticket-mcp MCP server to give Zed’s AI assistant direct access to your issues.

Configuration

Add to your Zed settings file (~/.config/zed/settings.json):
{
  "context_servers": {
    "ticket-mcp": {
      "command": "uvx",
      "args": ["ticket-mcp"],
      "env": {}
    }
  }
}
uv installed via: curl -LsSf https://astral.sh/uv/install.sh | shPython version 3.10 or higher (installed automatically by uv)Zed editor installed from zed.dev

Verify Installation

  1. Open the Agent Panel in Zed (sidebar)
  2. Go to the Agent Panel Settings
  3. Look for ticket-mcp in the list of context servers
  4. A green dot indicates the server is active and running
Zed reads settings on startup. If you edit settings.json while Zed is running, restart the editor to pick up changes.

Workflow Tips

Starting a Session

Ask Zed’s AI: “Run tk triage and show me what I should work on”

Creating Issues

Ask Zed’s AI: “Create a bug for the broken syntax highlighting, priority 1” Zed’s AI will run:
tk create "Fix broken syntax highlighting" -t bug -p 1

Closing Work

Ask Zed’s AI: “Resolve tk-abc123, the fix is committed”

Dependency Management

Ask Zed’s AI: “Show the dependency tree for tk-xyz789”

Project Health

Ask Zed’s AI: “Run tk insights and explain the critical path”

Zed-Specific Features

Both Zed and tk are built in Rust. If you’re working on Rust projects, tk’s instant performance (sub-10ms cached responses) matches Zed’s speed-first philosophy.
Zed’s Agent Panel provides a dedicated AI workspace where MCP tools are first-class citizens. The panel shows tool invocations inline with conversation context.
Ask Zed’s AI to analyze a tk issue and make changes across multiple files in a single Agent Panel session, with all edits visible as inline diffs.

Troubleshooting

Verify context_servers is at the top level of your settings.json (not nested under another key). Restart Zed after editing settings.
Hover over the status indicator for error details. Common causes: uvx not in PATH, Python version too old, or network issues during first-run package download.
Ensure you’re using the Agent Panel (not inline completions). MCP tools are only available through the Agent Panel’s AI assistant.
Enable caching with tk cache info to verify the analytics cache is working. Cached responses return in under 10ms.

Next Steps


Sources