ios-debugger-agent
Use XcodeBuildMCP to build, run, launch, and debug the current iOS project on a booted simulator. Trigger when asked to run an iOS app, interact with the simulator UI, inspect on-screen state, capture logs/console output, or diagnose runtime behavior using XcodeBuildMCP tools.
Install with Tessl CLI
npx tessl i github:Dimillian/Skills --skill ios-debugger-agent95
Does it follow best practices?
Validation for skill structure
Evaluation results
47%
37%Launch and Verify iOS App on Simulator
Core build-run workflow with session setup
list_sims used
0%
50%
Booted sim selected
0%
0%
session-set-defaults called
0%
50%
workspacePath in defaults
0%
62%
scheme in defaults
0%
62%
simulatorId in defaults
0%
25%
build_run_sim used
0%
50%
Log capture started
0%
40%
Log capture stopped + summarized
0%
20%
Screenshot taken
0%
50%
No boot attempt
100%
100%
Without context: $0.3200 · 1m 14s · 14 turns · 43 in / 3,809 out tokens
With context: $0.6150 · 2m 25s · 14 turns · 15 in / 3,006 out tokens
90%
82%Automate a Sign-In Flow on a Running iOS App
UI interaction and visual inspection
launch_app_sim used
0%
0%
describe_ui before tap
0%
100%
Tap uses id or label
33%
100%
No coordinates-only taps
0%
100%
type_text after focus
0%
100%
gesture for scroll
0%
100%
Screenshot taken
0%
100%
describe_ui after layout change
33%
100%
MCP tools preferred
0%
100%
Without context: $0.4423 · 2m 18s · 17 turns · 46 in / 8,256 out tokens
With context: $1.2349 · 4m 21s · 21 turns · 23 in / 3,872 out tokens
80%
64%Diagnose a Crash on Launch Using Console Output
Troubleshooting: bundle ID, console output, build failure
get_sim_app_path used
0%
75%
get_app_bundle_id used
0%
75%
Correct bundle ID order
0%
100%
captureConsole enabled
0%
100%
Relaunch for console
0%
50%
start_sim_log_cap with bundle ID
0%
80%
stop_sim_log_cap called
0%
87%
Build failure: asks about preferXcodebuild
14%
57%
No silent build retry
100%
100%
Without context: $0.8166 · 2m 59s · 36 turns · 5,376 in / 8,595 out tokens
With context: $0.9951 · 3m 54s · 32 turns · 951 in / 6,867 out tokens
- Evaluated
- Agent
- Claude Code
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.