Agent skill

Hopper Debugger

by steipete in steipete/agent-scripts

Hopper debugging: macOS/iOS binaries, ObjC/Swift symbols, dyld, LLDB.

MITAuto-check passedDevelopment

Install Hopper Debugger

skills CLI
$ npx skills add steipete/agent-scripts --skill hopper-debugger -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install steipete/agent-scripts hopper-debugger --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/steipete/agent-scripts.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/hopper-debugger .claude/skills/hopper-debugger && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
hopper-debugger
GitHub stars
7.3k
Token cost
~1.6k tokens
SKILL.md length
686 words
Files
2
Skills in repo
45
Repo updated
First seen
Licence
MIT

At a glance

Hopper debugging: macOS/iOS binaries, ObjC/Swift symbols, dyld, LLDB.

  • Works in 2 steps: Prefer Peter's pre-made exports at… → Extract fresh with…
  • Tasks that involve iOS development
  • SKILL.md covers Setup (one time), Call convention — the one that…, Licensing — do not misdiagnose and Opening documents, plus 4 more sections
  • Calls osascript

What it does

Hopper Debugger is an agent skill from steipete/agent-scripts. Hopper debugging: macOS/iOS binaries, ObjC/Swift symbols, dyld, LLDB.

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Development, covering iOS development. It works with macOS, Objective-C and iOS. The repository describes itself as: Scripts for agents, shared between my repositories. The licence is MIT.

When your agent uses it

  • Tasks that involve iOS development

Example prompts

  • “/hopper-debugger”

Workflow steps

2 steps, taken from the first numbered list in SKILL.md.

  1. Prefer Peter's pre-made exports at ~/Library/CloudStorage/Dropbox/Hopper/ — .hop documents plus .m pseudo-code dumps for AppKit…
  2. Extract fresh with /usr/lib/dsc_extractor.bundle (present on macOS; extracts all dylibs, multi-GB, slow).

What it can do on your machine

Read from SKILL.md and the folder at commit 79150cf. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • osascript

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Hopper Debugger loads about 1.6k tokens when it runs. Until then it costs about 21 tokens; SKILL.md has 686 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~21
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from steipete/agent-scripts at commit 79150cf, republished under its MIT licence (© steipete). 686 words, ~1,648 tokens.

Download SKILL.mdSave it as .claude/skills/hopper-debugger/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
hopper-debugger
description
Hopper debugging: macOS/iOS binaries, ObjC/Swift symbols, dyld, LLDB.

Hopper Debugger

Use Hopper through mcporter as a queryable disassembler, then combine the result with local source, LLDB, logs, and focused repros.

Setup (one time)

Hopper 6.0+ ships its own MCP server. Do not install a third-party one.

bash
mcporter config add hopper --scope home \
  --command "/Applications/Hopper Disassembler.app/Contents/MacOS/HopperMCPServer" \
  --description "Hopper Disassembler built-in MCP server (stdio)"

--scope home is required. The default scope is project, which writes config/mcporter.json into whatever repo you are standing in (untracked repo dirt, lost with the worktree).

Verify:

bash
MCPORTER_LIST_TIMEOUT=25000 timeout 40 mcporter list hopper --brief

Call convention — the one that bites

Always pass arguments with --args. Never --params. mcporter accepts an unknown --params flag silently, drops the payload, and the call arrives with no arguments. Hopper then answers Document not found., which reads like a licensing or state problem and is not.

bash
# WRONG — arguments silently dropped, fails with "Document not found."
mcporter call hopper.list_segments --params '{"document":"AppKit"}'

# RIGHT
mcporter call hopper.list_segments --args '{"document":"AppKit"}'

Related: the server does not fall back to the current document. A call with no document argument fails even when current_document returns a valid name. Pass document on every document-scoped call.

Document names come from list_documents and carry no .hop extension (window title AppKit.hop → document name AppKit).

If a call still fails, drive the server directly over stdio to see the raw JSON-RPC — this bypasses mcporter entirely and isolates who is at fault:

bash
"/Applications/Hopper Disassembler.app/Contents/MacOS/HopperMCPServer" <<'EOF'
{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"p","version":"1"}}}
{"jsonrpc":"2.0","method":"notifications/initialized"}
{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"list_documents","arguments":{}}}
EOF

Licensing — do not misdiagnose

Hopper 6.x still uses the bundle id and preferences domain com.cryptic-apps.hopper-web-4. The absence of a hopper-web-6 domain does not mean the app is unlicensed. Check the real state in the About panel (Hopper Disassembler > About Hopper Disassembler); it prints the licensee, order id, and update-plan expiry. There is no Register/License menu item in Hopper 6.

Peter's license: order HOP140213-7833-95831, updates through 2027-05-17. The .hopperLicense file is in 1Password (Molty vault, document item "Hopper Disassembler License (HOP140213-7833-95831)"); load $one-password to retrieve it. Opening a .hopperLicense file with Hopper does not register it — Hopper disassembles it as a document.

Opening documents

bash
open -a "Hopper Disassembler" /path/to/Binary

Small binaries import with no dialog. Large frameworks take minutes; poll instead of sleeping:

bash
until mcporter call hopper.list_documents --output json 2>/dev/null | grep -qi "appkit"; do sleep 10; done

Dismiss any first-open dialog with an Accessibility press, never a synthetic click — clicking moves Peter's physical pointer:

bash
osascript -e 'tell application "System Events" to tell process "Hopper Disassembler" to perform action "AXPress" of (button 1 of window 1)'

Apple frameworks

Apple frameworks live in the dyld shared cache, not on disk. Two routes:

  1. Prefer Peter's pre-made exports at ~/Library/CloudStorage/Dropbox/Hopper/ — .hop documents plus .m pseudo-code dumps for AppKit, AccessibilityKit, and others. Grepping the .m is often faster than any MCP round-trip (AppKit.m is 244 MB). Check provenance before trusting them for version work. These are snapshots; as of 2026-08 they predate macOS 26.6 and 27.0, and their class/method inventory differs from both live runtimes. Good for structure, unreliable for OS-version diffing.
  2. Extract fresh with /usr/lib/dsc_extractor.bundle (present on macOS; extracts all dylibs, multi-GB, slow).
Show full SKILL.md (271 more words)Show less

Query workflow

  1. Start from the local source path or runtime symbol you are trying to explain.
  2. Find the symbol, then inspect one small target at a time:
bash
mcporter call hopper.search_procedures --args '{"document":"AppKit","pattern":"addCursorRect"}' --output json
mcporter call hopper.procedure_pseudo_code --args '{"document":"AppKit","procedure":"0x185475b2c"}' --output json

procedure accepts a symbol name or a hex address. Other useful tools: list_documents, current_document, set_current_document, list_segments, list_procedures, list_strings, search_strings, procedure_info, procedure_address, current_procedure.

  1. Summarize the relevant control flow; do not paste large decompilations.
  2. Validate the hypothesis with LLDB/logging/repro before editing app code.

Pairing with runtime evidence

Disassembly tells you which store a value lands in; only the runtime tells you whether it got there. Read the pseudo-code first to learn which ivar/collection the API actually writes to, then read that exact store at runtime with class_copyIvarList + object_getIvar + perform. Instrumenting the wrong (legacy) path is the classic time sink: on modern AppKit, -[NSWindow _addCursorRect:cursor:forView:] is dead code, and cursor rects are stored in _NSTrackingAreaAKViewHelper's cursorAreas set.

Always run the same probe on a second machine at a different OS version before concluding "regression". Several no-op probes look identical on a known-good OS and a known-broken one; a control run is what tells you the probe is measuring nothing. See $remote-mac for the fleet and codexbar-ui-verification-quirks memory for the cursor-measurement harness.

Failure handling

  • Wrap Hopper calls with timeout; a modal or import can leave the transport stuck.
  • Do not send concurrent Hopper MCP requests during import.
  • Connection closed usually means Hopper is not running or is showing a modal. Check windows via System Events, then retry.
  • Document not found. almost always means missing arguments (see --args above), not a broken document.
  • If mcporter is wedged, prefer restarting its daemon over broad process kills:
bash
mcporter daemon stop && mcporter daemon start

© steipete, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/hopper-debugger of steipete/agent-scripts.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit 79150cf

Compare with similar skills

Hopper Debugger next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Hopper Debugger compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Hopper Debugger this skillsteipete/agent-scripts7.3k—~1.6kAutomated safety check: PassMIT
Core Data ExpertAvdLee/Core-Data-Agent-Skill314—~1.2kAutomated safety check: PassMIT
Native App Performanceharperreed/dotfiles3344 repos~538Automated safety check: PassNone
macOS Reversezhaoxuya520/reverse-skill40k2 repos~366Automated safety check: PassMIT
Apple Crash Log .NET Symbolicationdotnet/skills5.6k1 repos~2.4kAutomated safety check: PassMIT
Releasevaayne/mori303—~1.2kAutomated safety check: PassMIT

Similar skills

  • Core Data Expert

    AvdLee/Core-Data-Agent-Skill

    Expert Core Data guidance (iOS/macOS): stack setup, fetch requests & NSFetchedResultsController, saving/merge conflicts, threading & Swift Concurrency, batch operations & persistent history…

    314 GitHub stars~1.2k tokensUpdated 23 days ago
    DevelopmentAuto-check passed
  • Native App Performance

    harperreed/dotfiles

    Native macOS/iOS app performance profiling via xctrace/Time Profiler and CLI-only analysis of Instruments traces.

    334 GitHub starsUsed in 4 repos~538 tokens
    DevelopmentAuto-check passed
  • macOS Reverse

    zhaoxuya520/reverse-skill

    A skill your agent uses for authorized macOS and Mach-O reverse engineering including codesign, Objective-C/Swift recovery, endpoint security surfaces, and Apple platform malware analysis.

    40k GitHub starsUsed in 2 repos~366 tokens
    SecurityAuto-check passed
  • Official

    Resolves .NET runtime frames in Apple .ips crash logs to function names, source files and line numbers using dSYM symbols, atos and the Microsoft symbol server.

    5.6k GitHub starsUsed in 1 repo~2.4k tokens
    MobileAuto-check passed
  • Release

    vaayne/mori

    Release workflow for Mori macOS workspace terminal and MoriRemote iOS app.

    303 GitHub stars~1.2k tokensUpdated 2 mo ago
    MobileAuto-check passed
  • Swift Expert

    Jeffallan/claude-skills

    Builds Swift apps for Apple platforms with SwiftUI, protocol-oriented design, async/await, actors and Sendable checks, verified with swift build and swift test.

    12k GitHub stars~1.5k tokensUpdated 4 days ago
    MobileAuto-check passed

More from steipete/agent-scripts

All 45 skills in this repo
  • Mac Fleet Maintenance

    steipete/agent-scripts

    Inventories and maintains a fleet of Macs from a desired-state file: package updates, repo and Xcode sync, and disk, backup and security health reports.

    7.3k GitHub stars~4.8k tokensUpdated 3 days ago
    Auto-check passed
  • Agent Transcript for PRs

    steipete/agent-scripts

    Finds a coding agent's session log, trims and redacts it, and inserts it into a GitHub PR or issue only when the user has asked for a transcript.

    7.3k GitHub starsUsed in 1 repo~698 tokens
    Auto-check passed
  • Parallels macOS VM Lab

    steipete/agent-scripts

    Uses a clean Parallels macOS VM to test GUI automation, TCC permission prompts and screenshot tools like Peekaboo, verifying results from outside the guest.

    7.3k GitHub stars~1.8k tokensUpdated 3 days ago
    Auto-check passed
  • ClawSweeper Status

    steipete/agent-scripts

    Reports ClawSweeper's status with a bundled script: workflow health, active workers, queue health and recently merged, reviewed, commented and closed items.

    7.3k GitHub stars~972 tokensUpdated 3 days ago
    Auto-check passed
  • GitHub Project Triage

    steipete/agent-scripts

    Produces maintainer-facing triage cards for a project's GitHub issues and pull requests, each with its URL, risk, test state, blockers and a next action.

    7.3k GitHub stars~4k tokensUpdated 3 days ago
    Auto-check passed
  • Nano Banana Image Generation

    steipete/agent-scripts

    Generates and edits images with Google's Nano Banana 2 (Gemini 3.1 Flash Image) through a uv script, with a draft-then-final workflow and sizes from 512 to 4K.

    7.3k GitHub stars~1.4k tokensUpdated 3 days ago
    Auto-check passed

Questions about Hopper Debugger

What does Hopper Debugger do?

Hopper debugging: macOS/iOS binaries, ObjC/Swift symbols, dyld, LLDB. Hopper Debugger is an agent skill from steipete/agent-scripts. Hopper debugging: macOS/iOS binaries, ObjC/Swift symbols, dyld, LLDB.

When should I use Hopper Debugger?

Hopper Debugger fits situations like: tasks that involve iOS development.

How do I install Hopper Debugger in Claude Code?

Run `npx skills add steipete/agent-scripts --skill hopper-debugger -a claude-code`. Or copy the skill folder (skills/hopper-debugger in steipete/agent-scripts) into .claude/skills/hopper-debugger in your project. Claude Code loads it when a task matches its description.

How do I install Hopper Debugger in Codex?

Run `npx skills add steipete/agent-scripts --skill hopper-debugger -a codex`. Or copy the skill folder (skills/hopper-debugger in steipete/agent-scripts) into .agents/skills/hopper-debugger in your project. Codex loads it when a task matches its description.

Can I use Hopper Debugger in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add steipete/agent-scripts --skill hopper-debugger -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/hopper-debugger, .gemini/skills/hopper-debugger, .github/skills/hopper-debugger and .opencode/skills/hopper-debugger in your project.

What does Hopper Debugger need to run?

Going by SKILL.md and its folder, Hopper Debugger needs the command-line tools its instructions call (osascript).

Does Hopper Debugger access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Hopper Debugger safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Hopper Debugger use?

Hopper Debugger is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Hopper Debugger use?

About 1.6k tokens (SKILL.md is roughly 6.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Hopper Debugger?

Skills that share tags, products or a category with Hopper Debugger: Core Data Expert (AvdLee/Core-Data-Agent-Skill, 314 stars), Native App Performance (harperreed/dotfiles, 334 stars), macOS Reverse (zhaoxuya520/reverse-skill, 40k stars) and Apple Crash Log .NET Symbolication (dotnet/skills, 5.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Hopper Debugger?

steipete (a GitHub user) maintains it in steipete/agent-scripts, which has 7,259 GitHub stars. The repository holds 45 skills in this directory. The repository was last updated on October 4, 2026.

Source: steipete/agent-scripts on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.