LLM detection, reads the code, reasons about intent.
Signature scanners ask “does this file match a known bad pattern?”. Guardian Gaze’s LLM scanner asks “does this code behave like it has unauthorised intent?”, analysing execution flow, behavioural context and obfuscation patterns to catch the polymorphic, multi-stage and context-dependent threats that signature-only scanners miss by design.
Availability: included from Pro upwards, via the Premium add-on · LLM scanning: reasoning about code intent
The plugin view: findings, severity and next steps inside the WordPress admin
AI Detection at a glance
Polymorphic backdoors
Code that rewrites its own structure on every infection, variable names randomised, control flow permuted, string constants base64’d differently each time. No two copies have the same hash, so signature scanners are blind. The LLM reads intent and catches them.
Multi-layer obfuscation
Nested encoding, base64_decode(rot13(hex2bin(...))), stacked multiple layers deep. Signature engines that don’t unwrap recursively miss the payload. The LLM recognises the pattern of "code that exists only to hide other code" and flags it.
Database-resident malware
Payloads stored in wp_options, wp_posts or transients, invisible to file scanners. The LLM analyses suspicious DB content alongside file scans. See the database scanning page for detail.
Why signatures run out of road
Every signature-based scanner is answering one question: have we seen this exact thing before? When the answer is yes, it is the best tool available, fast, cheap and essentially never wrong. When the answer is no, it has nothing to say.
That limitation used to be tolerable, because writing new malware took effort and campaigns reused code for months. It is much less tolerable now. Generating a functionally identical backdoor with a different shape is close to free, which means the gap between a novel sample and a catalogued one has widened into the place where most successful compromises live.
This is the specific situation people describe when they say four scanners called the site clean and it was not. The scanners were not broken. They were asked whether the code matched something known, and it honestly did not.
Asking what the code is written to accomplish
The LLM layer asks a different question: what does this code do, and is that something a legitimate plugin would need to do? A function that reads a value from the request, decodes it, and executes the result is a remote shell. That is true regardless of what the variables are called, how many layers of encoding it hides behind, whether it has ever been catalogued, or how convincingly it has been made to look like part of a caching library.
The same reasoning also reduces false positives, which is the half of the story heuristic scanners struggle with. A minifier's output is dense and unreadable but has no interest in your request parameters. An obfuscated licence check phones a vendor's server and validates a key. Both look alarming to a heuristic and neither is a backdoor, and a layer that reasons about purpose can say so.
Every finding comes with an explanation written for a person rather than an analyst: what the code does, why it was flagged, what an attacker would use it for, and what to do next. A file path and a CVSS score tell an experienced developer plenty and tell a site owner nothing, and at nine on a Sunday evening the person deciding is usually the site owner.
Why this is the Pro line, and what it does not do
The AI layer is what Pro is for. Everything beneath it (signature, heuristic, integrity and database scanning) is in the free plugin, because a scanner that cannot find catalogued malware without payment is not a scanner. Reasoning about intent is the expensive part, computationally and otherwise, and it is the part that handles what nothing else can.
It is not magic and it is not a replacement for the other layers. It runs on top of them, on the code they surface, which is why the free layers are not degraded to make Pro look better. A signature match is still the fastest correct answer available, and when one is available the AI layer is not the thing that found your problem.
It also does not remove the need for judgement on a genuinely ambiguous file. What it does is put the ambiguity in front of you in language you can act on, rather than leaving it silently unreported because it matched nothing.
AI Detection - common questions
Is AI detection available on the free plugin?
No. The LLM layer is Pro. The free plugin includes signature, heuristic, integrity and database scanning, which cover catalogued threats well.
Does my code get sent somewhere?
Scanning runs on our infrastructure rather than inside your server, so code and security metadata are transmitted for analysis. Customer personal data is not the subject of the scan. The privacy policy sets out exactly what is sent and how long it is kept.
Is it just ChatGPT reading my files?
No. The reasoning layer sits behind the other four detection layers and works on what they surface, with the plugin's own context about WordPress structure, file provenance and integrity state feeding into the assessment.
Does it produce false positives?
Fewer than heuristics alone, because reasoning about purpose distinguishes obfuscated-but-legitimate code from obfuscated-and-malicious code. It is not zero, and any scanner claiming zero is describing a scanner that misses things.
What if it flags something and I disagree?
Every finding explains its reasoning, so you can evaluate it rather than take it on trust. Quarantine lets you isolate a file reversibly instead of deleting it, which is the right move when you are unsure.
Works with
AI Based Scanner →
Reasoning, not just pattern matching.
Database Scanning →
Payloads hide in tables, not just files.
AI Reports →
Every finding, explained in plain English.
See it on your own site
The free scan takes under five minutes and tells you the truth.