mcpField
L
Loggerhead

Loggerhead

Local
getloggerhead.comData & Analytics0Updated 2w ago

About Loggerhead

Loggerhead is a Mac app and MCP server for developers who want coding agents to inspect local application telemetry. It stores OpenTelemetry data on the Mac and offers tools for querying logs, metrics, and traces, as well as managing telemetry lint rules.

Key features

  1. 01Query logs, metrics, and traces
  2. 02Inspect trace spans and lint findings
  3. 03Manage workspace lint rules
  4. 04Mark findings fixed after code changes
  5. 05Local telemetry storage and loopback access

Use cases

  1. 01Investigate errors in a running application
  2. 02Inspect metrics and trace spans while editing code
  3. 03Review noisy logs or telemetry lint findings
  4. 04Adjust lint rules for a workspace

MCP support

The MCP server runs inside the Loggerhead Mac app, and the loggerhead-mcp bridge connects MCP clients to it over loopback. It exposes telemetry queries, lint-rule controls, and a stats resource.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository
Packages
mcpbloggerhead-mcp.mcpb

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Loggerhead MCP server

M8ven Score

Your coding agent reads your app's logs, metrics, and traces.

Loggerhead is a native Mac app that receives OpenTelemetry logs, metrics, and traces from the software you run on your own Mac, and stores them in a DuckDB file on your disk. The MCP server runs inside the app. This repository holds loggerhead-mcp, the small bridge that connects Claude, Cursor, Codex, Windsurf, or any other MCP client to it.

The tools read the telemetry. Seven tools change the lint rules of a workspace, each with a reason that the user sees and can undo in the app. One tool marks a finding fixed after a code change. The endpoint listens on loopback only. Nothing leaves the Mac.

The Loggerhead app must be installed and running. Download it first. The bridge exits when the app is not running. The free tier includes the MCP server.

Install

Claude Desktop

Download loggerhead-mcp.mcpb from the latest release and open it. Claude Desktop installs the extension.

Claude Code
claude mcp add loggerhead -- "/Applications/Loggerhead.app/Contents/Resources/loggerhead-mcp"
Cursor, Codex, Windsurf, and other stdio clients

Every stdio client takes the same two values. The .mcp.json in this repository is the file to copy:

{
  "mcpServers": {
    "loggerhead": {
      "command": "/Applications/Loggerhead.app/Contents/Resources/loggerhead-mcp"
    }
  }
}

The app itself writes this entry for you: open Settings › AI and click Add beside your tool.

The bridge finds the app by itself on a fixed ladder of local ports. You set no port, and it needs no permission to read the app's files. Full instructions: Set up the MCP server.

Tools

ToolPurpose
query_logsSearch log records by text, severity, and time.
query_metricsRead time-series metrics by name and time range.
query_tracesFind traces by service, operation, duration, and time.
get_traceRead every span of one trace.
list_environmentsList every workspace and environment on this Mac, with the receiver ports.
get_receiver_infoReport the effective OTLP ports and the addresses to send telemetry to.
query_improvementsList the telemetry-lint findings: noisy logs, unbounded labels, secrets, lost trace context.
list_lint_rulesList the lint rules of a workspace with their state, thresholds, exceptions, ignored findings, and recent changes.
query_ai_findingsList the findings of the last AI analysis run. It does not start a run.
set_lint_rule_enabledSwitch a lint rule on or off for a workspace, with a reason.
set_lint_thresholdSet one threshold of a lint rule for a workspace, with a reason.
reset_lint_thresholdPut one threshold of a lint rule back to its default, with a reason.
add_lint_exceptionHide the findings of one lint rule for a service, metric, attribute key, operation, or log message, with a reason.
remove_lint_exceptionRemove one lint rule exception, with a reason.
ignore_findingIgnore one lint finding by its id, with a reason.
unignore_findingStop the ignore of one lint finding, with a reason.
mark_finding_fixedMark one lint finding fixed after a code change, so the app judges it by the telemetry after the fix.

One resource, loggerhead://stats, reports the total counts of logs, metrics, and spans. Every parameter is in the tools reference.

Privacy Policy

The bridge and the server read telemetry from the Loggerhead app on the same Mac and hand it to the MCP client you connected. They collect nothing about you, they store nothing of their own, they share nothing with a third party, and they open no network connection except to 127.0.0.1. Data retention is the app's: the telemetry lives in a DuckDB file on your disk for as long as the app's retention setting keeps it, and you can delete the file. The full policy of the publisher, Rubics Code, is at https://getloggerhead.com/privacy. Contact: [email protected].

Registry

The registry name of this server is com.getloggerhead/loggerhead.

  • MCP Registry name: mcp-name: com.getloggerhead/loggerhead

Licence

The loggerhead-mcp bridge and the files in this repository are under the MIT licence.

The bridge contains open-source Swift packages. The .mcpb bundle has Acknowledgements.txt with the licence text and the notice text of each package.

The bridge connects to Loggerhead, a proprietary Mac app by Rubics Code (ABN 40 440 756 434), Australia. The app has a free tier, and the source of the app is not public. The Loggerhead terms of service apply to the app.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.