Tracebug
Local MCP server that lets coding agents read TraceBug browser bug reports
- Pricing
- Open source
- Platforms
- Chrome extension, firefox-extension, CLI
- Runs
- Locally
- Version
- 1.13.0
- Listing updated
- 3 days ago
About Tracebug
Tracebug is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.
A TraceBug report is a single offline .html file holding clicks and typing, console errors, network requests, annotated screenshots, navigation and environment details. This MCP server runs over stdio on your machine, finds exported reports in Downloads and lets Claude Code, Cursor, Windsurf or VS Code investigate from that evidence.
The workflow is: capture a bug in the browser extension, export the replay, add the server once with a single claude mcp add command, then ask your agent to debug it. Nothing is uploaded and there is no backend.
The same capture can also be copied as a GitHub issue, Jira markup, PDF or a Playwright test, or sent to GitHub, Linear and Slack with your own token after a prompt.
Install
Run in a terminal
claude mcp add tracebug -- npx -y tracebugAdd to claude_desktop_config.json
{
"mcpServers": {
"tracebug": {
"command": "npx",
"args": [
"-y",
"tracebug"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"tracebug": {
"command": "npx",
"args": [
"-y",
"tracebug"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"tracebug": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"tracebug"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"tracebug": {
"command": "npx",
"args": [
"-y",
"tracebug"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"tracebug": {
"command": "npx",
"args": [
"-y",
"tracebug"
]
}
}
}What it does
- Offline report format
- One self-contained file with replay, console, network and screenshots.
- Local stdio server
- Runs on your machine and uploads nothing.
- Auto-discovery
- Finds exported reports in the Downloads folder.
- Export options
- Issue Markdown, Jira markup, PDF and a Playwright test.
- Browser capture
- Chrome and Firefox extension plus an npm SDK.
Good for
- Hand an agent a reproducible browser bug
- Debug a failed request from captured network data
- Turn a bug into a GitHub or Linear issue
- Share a bug with someone who has no account
Pricing
Open source
Free and MIT; paid cloud add-ons may come later
Prices change. Check tracebug.dev before you buy.
Tracebug FAQ
Is it free?
Yes. Local capture, exports and the MCP server are free and MIT licensed; optional cloud features may come later as paid add-ons.
Does data leave my machine?
The MCP server runs locally and the pricing page says reports live on your machine.
Which agents can read reports?
Claude Code, Cursor, Windsurf and VS Code.
How do I add it?
Run claude mcp add -s user tracebug -- npx -y tracebug mcp.
Is Tracebug free?
Yes. Tracebug is open source, so you can use and self-host it for free.
How do I add Tracebug to Claude Code?
Run this command in your terminal: claude mcp add tracebug -- npx -y tracebug
What are the best alternatives to Tracebug?
The closest alternatives to Tracebug are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to Tracebug

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools