More screenshots
Works with the AI Agents you already use
Reverse Engineer Anything: Evidence-Disciplined Investigation
Investigate shipped software artifacts with evidence discipline — binaries, packages, firmware, app bundles — to answer what a program does, how versions differ, or whether a behavior claim holds.
$5
Reverse Engineer Anything: Evidence-Disciplined Investigation
Example session with this skill installed
A vendor claims version 2.0 removed a telemetry call. Verify it from the shipped builds: compare 1.9 and 2.0 and prove the call is gone — or find it.
- Read your context and instructions
- Compiled the reverse engineer anything
Routed first: two shipped builds, 1.9 and 2.0, same platform, both obtained from the vendor's own distribution — the comparison is only valid on matched artifacts. The ledger opened with the claim in column one: 'telemetry removed in 2.0'. Evidence went in column two as it was found; gaps in column three. Stage one compared the file lists and sizes: 2.0 is smaller by one library, and the telemetry library is absent from the package. Stage two checked the strings: the telemetry endpoint address appears in 1.9 and nowhere in 2.0. Stage three went deeper: the call sites in 1.9 that invoked the telemetry function are gone in 2.0, and the two remaining network calls go to the update endpoint, verified by address. One gap stayed open: the 2.0 build's update check sends a version string, which is a small data flow, not telemetry — recorded as a gap with its own evidence, not waved away. Verdict: the claim holds for the telemetry call, with the version-string flow named as the remaining observation. The ledger, the addresses, and the file hashes are filed, so anyone can reproduce the comparison. The vendor was shown the ledger, not just the verdict, and the version-string flow is now on their roadmap to disclose — the investigation changed the product, not just the argument.
Connects securely to your tools. The creator never sees your data.
About this skill
Some questions can only be answered from the artifact itself, because the source is gone, the vendor will not say, or the claim needs checking against the thing that actually ships. This skill investigates shipped software artifacts — binaries, packages, firmware images, app bundles — with evidence discipline: route the target, keep a three-column ledger of claims, evidence, and gaps, deepen in stages, compare versions, and close the loop with a finding that can be reproduced. It applies to a narrow set of questions: what a program does, how two versions differ, whether a behavior claim holds. When the question can be answered from documentation or source, the skill says so and stops — reversing what you could have read is wasted work. Authorization is part of the method: the target, the scope, and the permission are written down before the investigation starts.
It answers questions from artifacts; when documentation or source can answer the question, the skill stops and says so. An investigation is bounded by the artifacts supplied — a missing build or a stripped binary limits what can be proven, and the ledger records that. Authorization and scope are required up front; the skill does not investigate targets you have no right to examine.
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
Permissions
No special permissions declared or detected