More screenshots
Works with the AI Agents you already use
macOS Reverse: Analyze Mach-O Binaries and Behavior
Reverse-engineer macOS software with authorization: verify signatures, map Mach-O structure, recover Obj-C and Swift symbols, observe runtime behavior.
$5
macOS Reverse: Analyze Mach-O Binaries and Behavior
Example session with this skill installed
A client suspects a macOS app is phoning home. Triage the binary with authorization and tell me what it does, with evidence I can reproduce.
- Read your context and instructions
- Compiled the macos reverse
Authorization: the engagement letter names the app and the client's own machine; the triage runs on a copy, never the live install. Identity: codesign verification shows a valid developer signature and notarization — the binary is who it claims to be, so the question becomes what it does, not whether it is impersonating something. The entitlements list includes network-client and a keychain-access group, both noted before any code is read. Structure: otool shows the load commands and linked frameworks — Foundation, Network, and a third-party analytics framework the client did not know about. The analytics framework is the first lead, not a verdict. Code: class-dump recovers the Objective-C interfaces, and the Swift symbols resolve through the metadata — one class named for telemetry with methods for event batching and upload scheduling. Strings extraction finds two endpoint hostnames, both belonging to the analytics vendor. Behavior: ran under lldb with a breakpoint on the upload method and watched it fire 90 seconds after launch, sending a batch of event names — no file contents, no keystrokes, no credentials in the captured traffic. Frida confirmed the same call path on a second run. Handoff: the report lists every command run, the outputs, and the two hostnames, with the verdict stated plainly — the app phones home to its analytics vendor with event names only, which matches the vendor's documented behavior, and the client decides whether that is acceptable. Nothing in the traffic justified a stronger claim.
Connects securely to your tools. The creator never sees your data.
About this skill
macOS binaries hide their intent behind code signatures, load commands, and two languages' worth of runtime metadata — and this skill works through them in a fixed triage order so you never start by reading disassembly you do not need. It covers verifying code signatures and notarization first, mapping Mach-O load commands and dependencies, recovering Objective-C and Swift symbols, and observing runtime behavior with lldb and Frida. The handoff notes section is deliberate: a reversal that cannot be reproduced from its notes is a story, not evidence. iOS targets are routed to mobile analysis instead, because the platform differences are real.
Authorized analysis only: your own software, a client engagement, or a scoped assessment. It is not a tool for cracking or redistributing software you do not own.
iOS targets are out of scope here — the platform differences are real, and they belong to mobile analysis.
Heavily obfuscated or packed binaries may resist static recovery; the skill shows how to pivot to runtime observation when that happens.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
Requires