What is XcodeTraceMCP CLAUDE.md?

Repository instructions for an MCP server that connects coding assistants to Xcode Instruments' command-line profiling tools.

Guiding changes to the TypeScript library and MCP server that record, export, parse, symbolicate, and analyze performance traces from Apple apps.

They explain the project structure, security rules, and limits of the profiling data, so an agent can work with the code without treating it as a replacement for the Instruments app.

How jamesrochabrun/XcodeTraceMCP configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to jamesrochabrun/XcodeTraceMCP. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/jamesrochabrun/XcodeTraceMCP/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/jamesrochabrun/XcodeTraceMCP

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/jamesrochabrun/xcodetracemcp/claude-md/github.svg)](https://agentmods.dev/instructions/jamesrochabrun/xcodetracemcp/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/jamesrochabrun/xcodetracemcp/claude-md"><img src="https://agentmods.dev/badge/instructions/jamesrochabrun/xcodetracemcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/jamesrochabrun/xcodetracemcp/claude-md"><img src="https://agentmods.dev/badge/instructions/jamesrochabrun/xcodetracemcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 3,112 This file is loaded in full into every session.

When invoked 3,112 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin 98% copy Near-identical to another mod in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.03112	$0.03112
Opus 5.5	$0.01245	$0.01245
Sonnet 5.5	$0.00622	$0.00622
Haiku 4.5	$0.00311	$0.00311

Measured 2mo ago against content hash 28e9a370a4b0, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

XcodeTraceMCP CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

98% identical to XcodeTraceMCP AGENTS.md — 3 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

CLAUDE.md · 178 lines

How it starts

The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Guide

This repository contains a local Model Context Protocol (MCP) server for Xcode Instruments profiling. Treat it as an honest xcrun xctrace companion, not a replacement for Instruments.app. The server records, opens saved traces in Instruments.app by default, symbolicates, exports, parses, and analyzes what Apple exposes through xctrace, then reports unsupported or non-exportable data explicitly.

Project Architecture

The repo is a pnpm workspace with two packages:

packages/core: reusable TypeScript library for xcrun xctrace capability checks, recording/exporting, symbolication, trace parsing, analysis, recommendations, and Time Profiler comparisons.
packages/mcp-server: MCP stdio server that exposes the core library as assistant-callable tools.

High-level flow:

An MCP client calls a tool such as profile_running_app.
The MCP server validates arguments and delegates to @xctrace-analyzer/core.
Core runs xcrun xctrace using execFile argument arrays.
Core parses xctrace export --toc, TOC-driven XPath table XML, and HAR output when available.
Core returns typed analysis objects with support status and export attempts.
The MCP server formats those objects into Markdown, JSON, or both.

User-Facing Skill

This repo includes a bundled skill at skills/xctrace-profiler. Use it as the primary human-facing entry point for profiling, trace analysis, hangs, CPU bottlenecks, leaks, allocations, network, energy, startup slowness, and trace comparisons.

Users should be able to say simple prompts like:

Profile this app.
Find why this app is slow.
Profile this app for hangs.
Profile this app for hangs and tell me which of my code is responsible.
Check this build for leaks and allocation churn.
Analyze network activity.
Launch this app and profile startup hangs.
Analyze this trace.
Compare these two traces.

The skill should hide MCP JSON and tool names. It is the planning layer. It chooses between profile_running_app, track_running_app, analyze_trace, compare_traces, check_xctrace, list_templates, and list_devices, records or analyzes with outputFormat: "both" when diagnostics matter, reads Support Matrix / Export Diagnostics, and can rerun analyze_trace with timeRangeMs around the longest hang so ## Top User-Code Frames answers the app-owned code question.

Read the full file on GitHub · 178 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 178 lines · 3,112 tokens per session scan A 28e9a370a4b0

Subscribe to this mod's changes

XcodeTraceMCP CLAUDE.md is an instructions file published in the GitHub repository jamesrochabrun/XcodeTraceMCP (44 stars, last pushed 5mo ago), licensed MIT. It adds 3,112 tokens to every session, about $0.0124 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 98% identical to XcodeTraceMCP AGENTS.md, differing in 3 lines, and is treated as a copy.

Debugging

XcodeTraceMCP CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
XcodeTraceMCP CLAUDE.mdjamesrochabrun/XcodeTraceMCP	Instructions file	44	5mo ago	MIT	A	3,112	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,516	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	today	MIT	A	4,680	Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	2,022	GitHub Copilot

XcodeTraceMCP: Instructions file for Claude Code