Description
Someone just asked ChatGPT about your company. It made up the answer.
AI systems don’t show your website — they summarize it. They crawl your raw HTML (built for browsers, full of menus, scripts, and layout noise), guess what matters, and fill the gaps with whatever sounds plausible. The result: wrong product names, outdated pricing, invented features.
You can’t fix this with SEO. SEO controls how Google ranks your page. This is a different problem — AI models are replacing your page with a summary, and that summary is wrong.
LLM Override fixes the source of the problem.
When an AI crawler visits your site, LLM Override intercepts the request before WordPress renders any HTML and responds with a clean, structured document containing exactly your verified facts — nothing more, nothing less.
Your website stays the same. Your visitors see no difference. But when ChatGPT, Claude, or Perplexity talk about your brand, they work from your facts instead of guessing.
What actually happens (in plain English)
- An AI bot visits your page.
- LLM Override detects it and intercepts the request — before your theme even loads.
- Instead of messy HTML, the bot receives a clean document with your real content, your official terminology, and your verified company facts.
- Your human visitors? They see your normal website. Nothing changes for them. Nothing.
That’s it. No new URLs, no parallel site, no content to maintain separately.
What this solves (and what it doesn’t)
This is NOT a tool to “get mentioned by ChatGPT.” No tool can guarantee that.
This IS infrastructure that ensures: when AI systems do mention your brand, they do it accurately — using your real product names, your real pricing, your real company description. Not a hallucinated version.
Think of it this way: a sitemap tells an AI where your pages are. LLM Override controls what the AI actually reads when it gets there.
Who needs this
Ask ChatGPT about your company right now. Seriously — go do it.
If the answer is accurate, complete, and doesn’t mention your competitor in the same breath, you’re in the minority. For most B2B companies, the AI either gets the facts wrong, describes what you do in vague generic terms, or — worse — recommends a competitor when someone asks specifically about your category.
This isn’t a niche problem. It’s happening to every B2B company whose potential clients are using AI to research vendors, compare solutions, or shortlist providers. And that’s most of them, right now.
LLM Override is for any business that wants to control what AI says about them — instead of hoping it guesses right.
The complete free engine
LLM Override Free is not a teaser. It is a full, standalone system.
AI Interception
Detects 58 known AI crawlers (GPTBot, ClaudeBot, PerplexityBot, and 55 others) and serves them clean, structured content instead of raw HTML. Works automatically on every page and post type.
Terminology Standardization
Define find-and-replace rules that normalize outdated or incorrect terms across every AI payload. Renamed a product? Changed a pricing tier? Map the old name to the new one — globally, automatically.
Site Manifest
Write 3–5 sentences of verified company facts. These are included in every AI interaction with your site — anchoring your identity before the AI reads any page content.
llms.txt Compliance
Auto-generates a dynamic /llms.txt manifest (the emerging standard for AI discoverability) that always reflects your latest published content. Also generates /llms-full.txt with content previews. Both endpoints respect your existing SEO noindex rules from Yoast, Rank Math, SEOPress, or AIOSEO.
Content Faithfulness Score
Measures the parity between what your human visitors see and what AI bots receive — so you can prove there’s no “cloaking” or content manipulation.
“View as AI” verification
One-click button in the WordPress admin bar. See exactly what any AI crawler receives from any page on your site. No guessing.
Shadow Analytics
Track which AI bots are visiting, which pages they’re reading, and how often — with GDPR-compliant IP hashing. No data sent to external servers. Ever.
Developer hooks
14 documented filters and actions for extending behavior without modifying core plugin files. Full OOP architecture with Composer autoloading.
Zero impact on your existing site
This matters enough to say it clearly:
- Your theme renders exactly as it does today
- Your page speed is unaffected
- Your SEO rankings are untouched (AI responses carry a
noindexheader — Google never sees them) - Your caching plugins keep working normally
- Your existing plugins keep working normally
LLM Override operates on a separate delivery channel that human browsers never trigger.
Works everywhere WordPress runs
No filesystem writes. No server configuration. No special hosting requirements. Uses only the WordPress Options API, Transients API, and Rewrite Rules.
Verified on: WordPress VIP, WP Engine, Kinsta, SiteGround, Cloudways, Pressable, and standard Linux VPS environments. Compatible with Gutenberg, Elementor, Divi, WPBakery, and Beaver Builder.
LLM Override Pro — for large sites and agencies
The free version is a complete engine. Pro adds scale.
- M2M Precision Parser — AI-powered per-page structural translation (via your own API key — OpenAI, Anthropic, DeepSeek, or OpenRouter). Strict 1:1 content parsing, no rewriting, no summarizing.
- Batch Compilation Engine — compile your entire site’s AI payloads in the background via Action Scheduler. No timeouts, no manual work.
- Full GEO Analytics — granular telemetry: which bots, which pages, which terms were normalized, Content Faithfulness scoring per URL.
- Autopilot llms.txt — AI-drafted Master Fact Manifest grounded strictly in your actual page content.
- Agency MCP Server — Model Context Protocol endpoint for remote agent orchestration across a fleet of client sites.
GEO Compliance
LLM Override follows the principle of Content Faithfulness: everything served to
AI bots is a faithful representation of your visible web content, optimized for
machine consumption but never altered in substance.
- Content is converted from HTML to Markdown without semantic changes
- No content is added or removed from individual page responses
- Structured data (JSON-LD) is extracted directly from visible page content
- Full transparency via X-Content-Processing HTTP headers
- Content Faithfulness Score tracks parity between HTML and Markdown outputs
Screenshots

Dashboard — engine status, bot activity counter, intercepted URLs, and health checklist at a glance. 
“Before vs. After” — see the raw HTML an AI would receive without the plugin versus the clean structured payload it receives with it. 
Content Rules — Terminology Standardization and Site Manifest configuration in one panel. 
llms.txt Config — one-click regeneration, real-time stats, endpoint preview. 
Per-post metabox — exclude any page from AI manifests or manually override its AI payload.
Installation
- Upload the
llm-overridefolder to/wp-content/plugins/. - Activate via the Plugins menu in WordPress.
- Go to LLM Override Dashboard — your interception engine is active immediately.
- (Recommended) Write your Site Manifest under LLM Override Content Rules — 3–5 sentences of verified company facts.
- (Optional) Set Terminology Standardization rules and configure your
/llms.txtpost types under LLM Override llms.txt Config.
No configuration is required for basic operation. The engine works out of the box.
FAQ
-
ChatGPT already mentions my brand. Why do I need this?
-
Because you can’t verify whether what it says is accurate. ChatGPT synthesizes answers from whatever HTML it can parse — including outdated blog posts, legacy product names, and content mixed with navigation menus and JavaScript. LLM Override gives it a clean, verified document instead. The goal isn’t more mentions. It’s accurate mentions.
-
Does this change anything for my human visitors?
-
Nothing. The AI delivery channel only activates for recognized AI crawlers. Your visitors always receive your normal WordPress theme. There’s no redirect, no separate URL, no parallel site.
-
Does it affect my Google rankings?
-
No. Every AI response includes an
X-Robots-Tag: noindexheader. Google never sees the AI channel. Your HTML pages continue to be indexed exactly as they are today. -
What’s the difference between this and llms.txt?
-
llms.txt is a directory — a list of your important URLs. LLM Override generates that directory automatically. But it also goes further: when an AI bot visits those URLs, it receives structured, verified content instead of raw HTML. The directory is the door. LLM Override controls what’s inside the room.
-
What’s the difference between this and a sitemap?
-
A sitemap tells AI where your URLs are. LLM Override controls what AI receives when it visits those URLs. A sitemap is a phone book. LLM Override is the actual conversation.
-
Does it work with my hosting provider?
-
Yes. LLM Override never writes files to your filesystem. It works on any WordPress hosting environment — including managed platforms with read-only filesystems like WordPress VIP, WP Engine, and Kinsta.
-
Does it conflict with caching plugins?
-
No. It detects WP Rocket, LiteSpeed Cache, W3 Total Cache, Cloudflare, Varnish, and FastCGI — and bypasses caching only for AI requests. Your human visitors continue to get cached pages normally.
-
Does it work with page builders?
-
Yes. Fully compatible with Gutenberg, Elementor, Divi, WPBakery, and Beaver Builder. The engine extracts content directly from the database before the theme renders.
-
What AI bots does it detect?
-
58 known crawlers across 4 categories: Training bots (data harvesting), Query bots (real-time AI answers from GPTBot, ClaudeBot, PerplexityBot), Discovery bots (site mapping), and Scraping bots (unclassified AI traffic).
-
Is it GDPR-compliant?
-
Yes. IP addresses are hashed with a daily-rotating salt — never stored in plain text. No data is transmitted to external servers.
-
Does it send my content to OpenAI or other AI companies?
-
No. The free version is entirely self-contained. It serves content to incoming AI crawlers — it never makes outgoing API calls. The Pro version optionally connects to AI APIs via keys you provide (BYOK), only when you explicitly trigger a compilation.
-
I’m a developer. What hooks are available?
-
14 documented hooks. Key filters:
llm_override_markdown_output(modify final payload),llm_override_yaml_frontmatter(extend metadata),llm_override_pre_convert_content(replace HTML before conversion),llm_override_llmstxt_entries(filter manifest URLs),llm_override_bot_user_agents(extend bot dictionary). Key actions:llm_override_bot_detected(fires on interception),llm_override_before_markdown_output/llm_override_after_markdown_output(lifecycle hooks). Full OOP architecture with Composer autoloading.
Reviews
There are no reviews for this plugin.
Contributors & Developers
“LLM Override” is open source software. The following people have contributed to this plugin.
ContributorsTranslate “LLM Override” into your language.
Interested in development?
Browse the code, check out the SVN repository, or subscribe to the development log by RSS.
Changelog
1.1.6
- Performance: Resolved critical AJAX timeout when regenerating llms.txt cache on sites with 1,000+ published pages. The admin stats engine now uses lightweight SQL COUNT queries (~2ms) instead of heavy WP_Query meta lookups that generated multiple LEFT JOINs on wp_postmeta.
- Performance: Optimized
/llms.txtand/llms-full.txtendpoint queries by disablingSQL_CALC_FOUND_ROWS(no_found_rows) and taxonomy cache preloading (update_post_term_cache), reducing memory usage and query time on large sites. - Performance: Fixed Dashboard KPI query that loaded all post IDs into memory unnecessarily. Now fetches only the count.
- Fix: Extended the safe shortcode stripping logic (introduced in 1.1.5 for the Content Pipeline) to all four remaining
strip_shortcodes()call sites in the llms.txt and llms-full.txt generators. Divi and WPBakery content is now preserved consistently across all endpoints. - Enhancement: llms.txt and llms-full.txt caches now auto-invalidate when any public post is published, updated, or trashed, eliminating up to 12 hours of stale content.
1.1.5
- Fix: Prevented destructive Markdown conversion when processing pages built with shortcode-heavy visual builders (Divi, WPBakery). The core pipeline now uses an intelligent recursive loop instead of
strip_shortcodes()to safely extract text payloads.
1.1.4
- Fix: Migrated all inline
<script>tags to usewp_register_script,wp_enqueue_script, andwp_add_inline_scriptAPIs, fully compliant with WordPress enqueue standards. - Fix: Resolved unescaped output variables in metabox template by wrapping all dynamic attributes in
esc_attr(). - Fix: Corrected JSON-LD script injection by removing unsafe
JSON_UNESCAPED_UNICODEandJSON_UNESCAPED_SLASHESflags fromwp_json_encode(), preventing potential</script>breakout. - Fix: Restructured
printfcalls in the llms.txt admin partial to usewp_kses()with explicit allowlists instead ofesc_html__()with raw HTML arguments. - Fix: Removed non-permitted binary files from distribution (
vendor/bin/html-to-markdown,vendor/league/html-to-markdown/bin/). - Fix: Eliminated duplicate Metabox instantiation that caused hooks to register twice.
- Tweak: Extracted Terminology Standardization repeater JavaScript into a dedicated external file (
admin/js/llm-override-admin-terminology.js), enqueued conditionally on the Content Rules page only. - Tweak: Yoast SEO dismiss handler now passes nonce via
wp_localize_script()instead of inline PHP. - Tweak: Added
phpcs:ignoreannotations with technical justifications for text/plain API endpoints whereesc_html()would corrupt Markdown output.
1.1.3
- Fix: Removed residual AJAX callback registrations pointing to non-existent methods, preventing a fatal error in certain configurations.
1.1.2
- Fix: Resolved a backend conflict by removing unused AJAX event listeners that could potentially trigger HTTP 500 errors in specific environments.
1.1.1
- Fix: Eliminated execution conflicts by removing non-functional code blocks.
- Performance: Restored ultra-lightweight architecture by ensuring all processes rely exclusively on the WordPress Options and Transients APIs.
1.1.0
- Feature: Terminology Standardization Engine — globally replaces legacy terms with official equivalents to ensure Content Faithfulness.
- Enhancement: Migrated global term filtering logic to comply with accurate Source Attribution guidelines.
1.0.5
- New: RAG JSON-LD Grounding Engine — injects semantic TechArticle schema markup into HTML
<head>containing M2M translated content for discovery bot ingestion. - Enhancement: Complete Content Pipeline refactoring. HTML-to-Markdown conversion centralized to prevent theme builder conflicts.
- Fix: Developer environments (localhost, IDE headless browsers) no longer trigger false positive bot interceptions.
1.0.4
- Fix: Excluded performance auditing tools (Lighthouse, GTmetrix, Pingdom) from bot detection.
- Fix: Extended SEO bot exclusions (AhrefsBot, SemrushBot, Applebot, DotBot, MJ12bot).
1.0.3
- Fix: Critical Indexing Hotfix — excluded Googlebot and Bingbot from AI interception to preserve normal SERP indexing.
1.0.2
- Fix: Changed Content-Type to text/plain for compatibility with strict AI URL ingesters (Google NotebookLM).
1.0.1
- New: Passive Yoast SEO Compatibility Checker with actionable dashboard alerts.
- Fix: Added missing Content-Type header to M2M payload response.
1.0.0
- Initial release.
- M2M Interception Engine with HTML-to-Markdown conversion.
- Site Manifest and Terminology Standardization.
- Dynamic
/llms.txtand/llms-full.txtendpoints. - Shadow Analytics Lite with 58-bot detection dictionary.
- SEO plugin synchronization (Yoast, Rank Math, SEOPress, AIOSEO).
- “View as AI” admin verification tool.
- 14 documented developer hooks.
- Full WordPress coding standards compliance: 0 Plugin Check errors, 0 warnings.
