explaingit

r00tify/crashdx

Analysis updated 2026-07-25

6SwiftAudience · developerComplexity · 3/5Setup · moderate

TLDR

A tool for Apple developers that analyzes iOS and macOS crash report files and provides a ranked list of likely root causes with supporting evidence, not just the line of code that failed.

Mindmap

mindmap
  root((repo))
    What it does
      Diagnoses crash causes
      Ranks likely reasons
      Shows supporting evidence
    Components
      Swift library
      Command line tool
      MCP server
    How it works
      Deterministic rules
      No machine learning
      Traceable evidence
    Use cases
      Debugging crashes
      Automated testing
      AI agent analysis
    Crash types handled
      Watchdog timeouts
      Memory pressure kills
      Stack overflows

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Diagnose the root cause of an iOS or macOS app crash from a .ips report.

USE CASE 2

Automate crash analysis in a testing pipeline using JSON output.

USE CASE 3

Let an AI agent analyze crash reports via the MCP server.

USE CASE 4

Build a custom crash analysis tool using the Swift library.

What is it built with?

SwiftMCP

How does it compare?

r00tify/crashdxasaptf/swift-language-modelsiamwilliamli/livetranscriber
Stars666
LanguageSwiftSwiftSwift
Setup difficultymoderatemoderatehard
Complexity3/54/53/5
Audiencedeveloperresearcherdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Requires building Swift components and likely has Apple platform dependencies.

In plain English

crashdx is a tool for Apple platform developers that reads crash report files and figures out why an app crashed, not just where it crashed. When an iOS or macOS app crashes, the system produces a file called an .ips report. Standard tools tell you which line of code was running when the crash happened. crashdx goes further: it applies a set of rules to the report data and produces a ranked list of likely causes, each backed by specific evidence from the file. The tool ships in three forms. There is a Swift library called CrashDXCore that other programs can build on, a command line tool called crashdx for running analyses directly, and an MCP server called crashdx-mcp that lets AI agents and automation systems call the same engine. All three use the same underlying logic, so a human running the command line tool and an automated pipeline get identical results. Two design choices stand out. First, every claim the engine makes can be traced back to the exact spot in the crash report that supports it. The tool does not ask you to trust its conclusions blindly. Second, when the evidence is ambiguous, it says so. Rather than forcing a single answer, it presents the competing possibilities, scores them, and explains what additional information would settle the question. The engine contains no machine learning or AI calls. It works through deterministic rules applied to verifiable facts. The README includes several real output examples. In one, the tool diagnoses an uncaught Objective-C exception by identifying specific frames in the stack trace that are telltale signs of that problem. In another, the tool returns an inconclusive verdict with two competing hypotheses for a memory access crash, suggesting specific debugging steps to distinguish between them. It also shows diagnoses for watchdog timeouts, memory pressure kills, stack overflows, and Swift runtime traps, each with its own evidence summary. The tool can output results as JSON for use in automated testing pipelines. The full README is longer than what was shown.

Copy-paste prompts

Prompt 1
Use crashdx to analyze this .ips crash report file and tell me the ranked list of likely causes with evidence.
Prompt 2
Write a script that uses the crashdx command line tool to analyze a folder of crash reports and save the JSON output for each.
Prompt 3
Explain how to integrate the crashdx-mcp server into an AI agent workflow to automatically diagnose app crashes.
Prompt 4
Show me how to use the CrashDXCore Swift library to build a custom tool that reads a crash report and extracts the evidence for a specific hypothesis.

Frequently asked questions

What is crashdx?

A tool for Apple developers that analyzes iOS and macOS crash report files and provides a ranked list of likely root causes with supporting evidence, not just the line of code that failed.

What language is crashdx written in?

Mainly Swift. The stack also includes Swift, MCP.

How hard is crashdx to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is crashdx for?

Mainly developer.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.