# Product updates Source: https://docs.foenix.ai/changelog New features, improvements, and fixes to Foenix.ai ## Agents: Worker replaces Architect * **Meet Worker.** The Architect agent is retired. On a heavy, long task the Orchestrator now decomposes it: splits the task into parts and calls a dedicated Worker sub-agent for each one. The result is noticeably higher accuracy and consistency across multi-step jobs. * **Verify is stricter and more honest.** It now checks whether the task was actually completed, not just whether it ran. * **Pause only pauses the schedule.** Pausing a Scheduled Agent stops its next timed runs. Sessions already in progress, including ones started with Run Now, keep going. ## Rollback: honest by design * **Rollback reports the outcome honestly.** If a change could not be undone, you find out right away instead of discovering it later. Every WordPress stack is different, and we keep extending rollback coverage to new configurations. * **Undo now covers long posts** with images and SEO meta, which previously could fall outside rollback. ## Make your site AI-ready * **llms.txt and similar files can now be created.** Ask Foenix to bring your site up to speed for AI crawlers and add llms.txt. ## Cloud Site, formerly Sandbox * **Sandbox is now Cloud Site.** Your sites are labeled Cloud Site or Self-hosted, and the demo one shows "Demo site expires in". * **500 bonus credits when you connect your own site.** One time, on the first site you connect. ## Live mode * **The current page URL is shown,** clickable, with long paths shortened. * **Links to your site's pages open in Live mode.** Session results now use three link styles so you can tell them apart: green for files created inside Foenix, orange for WP Admin, grey for external. ## πŸ›  Fixes * **Files** not reaching the chat, or opening with a 404. * **Responses occasionally switching language** mid-session. * **Plugin:** table creation on update, transaction errors during install. Update to the latest plugin version. * **Requests hanging** with no response until a page reload. ## New * **Pull-only plugin mode.** If your firewall blocks incoming requests, the plugin can now work in pull-only mode, so Foenix connects to sites that were previously unreachable. * **New Agent, your way.** The New Agent button now opens a choice: build the agent through a chat, or fill in the setup form yourself. * **Continue an agent session.** A finished agent's session can now be picked back up simply by writing a message into it. * **ZIP files are now readable.** Upload a plugin or theme archive and Foenix will unpack it and read the files inside. ## Faster and steadier * **Sandboxes launch noticeably faster.** They are now prepared in advance, with the plugin version kept current, so you spend less time waiting. * **Faster dashboard and project opening.** Fewer round trips to the server on every screen. ## Chat * **Hover a message to see when it was sent.** * **Improved the UI on mobile devices.** ## Fixes * **Critical:** sessions that could hang without completing are fixed, along with the credit accounting behind them. * **Buttons that needed several clicks** now respond the first time, and show progress when the server is slow. * **Rollback now reports honestly.** If a change could not be undone, you are told, instead of seeing a false success. ## Sandbox: relaunch anytime * **Launch a fresh 24-hour sandbox whenever you need one.** If your trial sandbox has expired, spin up a new one right from the dashboard. The button activates as soon as your previous sandbox winds down. * **Expired sandboxes are now cleaned up right on schedule.** * **WP Easy Migrate is now available as a direct download.** Install it on any WordPress site, not just inside a sandbox. [Guide and link in the docs](https://docs.foenix.ai/get-started/wp-easy-migrate). ## Scheduled Agents * **Memory for Scheduled Agents.** Scheduled Agents now retain knowledge of your WordPress site between runs. * **Cleaner responses:** stray internal reasoning no longer leaks into replies, and action formatting glitches are fixed. ## Quality of life * **Email when your task finishes.** Start a generation, walk away, and we'll let you know when it's done. * **Connection-loss handling.** If your internet drops mid-generation, you'll see a clear notification and your message returns to the input box, ready to resend. * **Google sign-in now runs in the same tab,** with no popup. ## Fixes * **Critical:** password reset was not working. Fixed. * **Images:** image generation and image analysis both back to normal. * **Plugin:** activation error when activating immediately after install is fixed. Connection issues for some sites are fixed. The Update button now takes you to WP Admin as expected. * **Generation pipeline:** a large batch of fixes and corrections in the places where errors occurred. ## Sandbox migration β€” as promised Two weeks ago we said our own migration plugin was coming. It's here: * **WP Easy Migrate** β€” our migration plugin, now preinstalled in every sandbox. Move an existing WordPress site (database + files) into a sandbox, or take what you built in a sandbox out to any WordPress site. ## Smoother start * **Straight into your workspace after signup** β€” no interstitial page, your sandbox spins up while you watch. * **Live launch status** instead of a generic spinner: "Spinning up WordPress… Installing the foenix plugin… Preparing your sandbox…" * **Email when your demo sandbox is ready** β€” even if you didn't wait around, you can come back with one click. * **WP Admin button** right next to your sandbox domain. * **Pin/unpin sandboxes** on the dashboard. ## Agents & models * **Standard mode is now Plus.** Lite runs on a mix of fast, lightweight models; Plus and Ultra use larger, more capable models. * **Plus and Ultra modes are now exclusive to paid plans.** * **Agents created from a chat now inherit that chat's model mode** β€” start in Lite, and the agent runs in Lite too, image models included. * **Run Now** on an agent takes you straight to the created session. * **Multi-provider model routing** for faster, more reliable responses in Lite mode. * **Updated Prompt Library** and redesigned quick-start chips. ## πŸ›  Fixes * **Files:** improved reading of uploaded text files. File names with spaces handled properly. * **Sessions:** response arriving but the interface not updating β€” fixed. Infinite loading β€” fixed. "Thinking" indicator stuck after a task finished β€” fixed. Empty responses in Review + Lite β€” fixed. * **Plugin:** plugins being unexpectedly deactivated on connected sites β€” fixed. * **UI:** double-click needed on preview and pricing buttons β€” fixed. Zip file downloads β€” fixed. Wrong mode shown on a newly created agent β€” fixed. ## Sandbox: built end-to-end Sandbox went from early access to a real, stable feature this cycle: * **Sandboxes deploy fast** β€” without a long wait to get one running. * **One-click WP Admin access.** Links Foenix generates while working on your site (installing a plugin, opening a settings page) now log you in and take you straight there β€” no more dead links. * **Real-time deploy status.** Sandboxes report Deploying β†’ Live as they come up, and you can now launch multiple sandboxes in parallel. * **Transparent, correct pricing.** 5GB (12 credits) / 10GB (20 credits) / 25GB (30 credits), with disk sizes that actually match what you picked. * Free / Starter / Pro sandbox limits now enforced correctly across the board. ## Agent system: fully reworked * **Three model modes.** Lite (MiniMax M3), Standard (Gemini 3.5 Flash), Ultra (Claude Sonnet 4.6) β€” chosen per session, plus an independent image-model selector. * **Mode choice locks in per session** once a task starts, so it can't silently change mid-run. * **Orchestrator and Automation agents fully aligned** β€” scheduled/automated runs now behave exactly like the agents you configure manually. * **Stopping a task no longer loses context** for follow-up requests in the same session. Added a proper Stop button. * **Prompt-caching rework** that meaningfully cuts wasted token cost on longer sessions. * **New Prompt Library** and quick-start chips β€” one click to launch ready-made recipes (daily blog post with images, weekly SEO audit, and more). * Faster responses in Lite mode via priority processing for MiniMax M3 and other lightweight models. * Fixed token/cost accounting across our models β€” including a caching double-count bug and garbled Flash responses. * Fixed autonomous agent lists not filtering correctly by site. ## New * **Review and Publish.** Choose where the result goes: **Review** prepares files for you to check first, **Publish** applies changes to the live site at once. * **Redesigned auth pages** β€” sign-up, login, password reset, and confirmation. * **Dashboard tiles:** new sites now appear at the top. * **File previews** now render content directly in a modal (with a download button) for more file types. * Renamed **Vibe Coding β†’ Live** β€” opens a preview of your site where you can turn on **Visual Edit** mode. * **Universal links.** Out of credits mid-task? Agents can now point you straight to the pricing page or the buy-credits modal. ## πŸ›  Fixes * **Auth:** Google sign-in occasionally failing on first attempt β€” fixed. Free-plan bonus credits now correctly granted on Google sign-up. * **UI:** Long names now display correctly in your profile. Dashboard load-time fixes. General polish across profile, sessions, and mobile layouts. * **Plugin:** Connection errors now shown clearly on the plugin settings page. ## New * **Gemini 3.1 Pro** now available on production across all agents. * **Credit packs now available on Starter plan.** Previously limited to Pro only. * **Session titles** now generated from chat context for easier browser history navigation. ## πŸ›  Fixes * **Critical:** WP plugin returning empty response on code execution β€” fixed, further improvements in progress. βš™οΈ * **Critical:** Plugin not loading dashboard on first connect β€” fixed. * **Billing:** Complete rewrite of credit counting logic. * **Billing:** Overage no longer deducted from future purchases or renewals. * **Billing:** Credit pack expiry corrected to 6 months. * **Billing:** Upgrade/Downgrade blocked correctly when renewal not allowed. * **UI:** Interface locked during backend requests β€” fixed, monitoring ongoing. βš™οΈ * **UI:** Skeleton animation added to profile β€” prevents errors from clicking before data loads. * **UI:** Edit agent panel moved to bottom of session with next run time and status visible. * **UI:** Vibe Coding top bar layout fixed. ## New * **Gemini 3 Flash** added as a model option alongside Claude Sonnet 4.6. Choose per task β€” Flash for speed and cost, Sonnet for complex builds. Configure in **Settings β†’ Personality β†’ AI Pipeline Models**. * **Create Autonomous agents from any session.** While in Conversational mode, just describe the task and tell Foenix when and how often to repeat it β€” for example, *"every Monday at 9am, audit my SEO and email me the report"*. Foenix expands it into a full prompt and shows a **Schedule Task** card to confirm. * **Stop a running task at any time.** You can now interrupt a task before it finishes β€” useful when you spot it heading the wrong way or just want to cancel and try a different prompt. * **Per-agent model selection.** In your Profile, you can now choose which AI model each agent uses individually β€” pick a lighter, cheaper model for simple recurring tasks and a more powerful one for complex builds. ## πŸ›  Improvements * A batch of bug fixes across the platform. We're watching closely and shipping improvements continuously β€” keep the feedback coming. Foenix.ai is in active early development. We've been building for the past 6 months and this is our current feature set. ### What's available today * **Multi-Agent architecture** β€” Foenix operates as four specialised agents (Orchestrator, Architect, Coder, Verifier) with built-in retry-on-failure logic. See [Architecture](/core-concepts/architecture). * **Rollback** β€” undo file-level changes (in active development) * **Safety modes** β€” three levels (Safe, Balance, Hard) to control how strictly Foenix validates actions * **Multi-site dashboard** β€” manage multiple WordPress sites from one account ### Honest note This is an early version of the product. We recommend using Foenix always keeping backups at the hosting level. We're moving fast and shipping improvements weekly. [Join our Discord β†’](https://discord.gg/Rammm3ts5u) # Architecture Source: https://docs.foenix.ai/core-concepts/architecture How Foenix's multi-agent system processes your requests Foenix is built as a **multi-agent system (MAS)** - four specialised agents work together, each with a narrow role. This is what makes Foenix able to run autonomously without losing control of the outcome. ## The two environments Foenix has two parts that talk to each other over HTTPS: **Foenix Cloud** - where the agents live. They reason about your request, plan steps, generate code, and verify results. **Your WordPress site** - where the Foenix Connector plugin runs. It receives instructions from the Cloud, executes them on your server, and reports the result back. The Connector has admin-level access to WordPress but only acts on signed instructions from your account. ## The four agents Every request goes through up to four agents. Simple requests use one or two; complex ones use all four with multiple iterations. ### Orchestrator - the lead Receives your message, decides which execution path to take, and coordinates the other agents. It picks between paths depending on the request: a quick read-only answer, a single-step edit, a multi-step generation, an autonomous run, and so on. On a heavy, long task the Orchestrator decomposes the request: it splits the task into parts and calls a dedicated Worker sub-agent for each one. The Orchestrator is also responsible for the retry loop: if the Verifier returns a failure, the Orchestrator decides whether to try again with a new strategy or report the problem. ### Worker - the sub-agent Used for heavy tasks. Every Worker gets its own slice of the job, researches the context it needs (theme structure, custom post types, existing plugins, ACF fields, taxonomies), and hands the Coder a precise, scoped instruction. Because each part is handled by its own Worker, a ten-step task no longer depends on a single plan staying valid from step one to step ten. The result is noticeably higher accuracy and consistency on multi-step jobs. ### Coder - the executor Translates the instruction into PHP and runs it on your WordPress server through the Connector. The Coder doesn't make decisions about *what* to do - only *how* to do it. This separation is deliberate: the agent that decides isn't the agent that executes. The Coder reads freely but writes carefully. Every modification is wrapped in a transaction, logged before execution, and screened by the Safety Mode validator before it runs (see Safety Mode). ### Verifier - the checker Read-only access to your site. After the Coder runs, the Verifier checks whether the result actually matches the original request - does the page render? does the plugin activate without errors? did the meta description get added? It returns a structured pass/fail verdict. The Verifier checks whether the task was actually completed, not just whether the code ran. A plugin that activates but does nothing, or a page that renders but is missing the requested section, fails verification. If it fails, the Orchestrator gets the verdict and retries with a corrected plan. ## How a request flows You type a message, click an element, or a scheduled agent fires. The Orchestrator picks an execution path based on the request type and your current Safety Mode. For heavy tasks, the Orchestrator splits the request into parts and assigns each to a Worker sub-agent. The Coder runs PHP operations on your server, one at a time, each with a transaction ID. The Verifier checks the result. Pass β†’ done. Fail β†’ back to the Orchestrator for a retry. You see the result in your session. ## Why multi-agent A single LLM doing everything has two failure modes that matter at scale. **It can't reliably check itself.** The same model that wrote the code is biased toward seeing it as correct. **It can't recover from errors during long autonomous runs.** If a 10-step task fails on step 4, a single-agent system tends to either crash or hallucinate completion. Splitting roles solves both. The Verifier doesn't share the Coder's blind spots. The Orchestrator can detect a failed verdict and decide what to do next without needing the original context that produced the error. And because heavy tasks are split across Workers, a failure in one part is isolated: the Orchestrator retries that part, not the whole task. # Credits Source: https://docs.foenix.ai/core-concepts/credits How Foenix pricing works: credits, AI Modes (Lite, Plus, Ultra), plans, credit packs, and what real tasks cost. Foenix uses a credit-based system. Every action your session performs consumes credits - reading data, writing code, generating images, searching the web, and running agents. ## Free credits Every new account gets **500 free credits** - no credit card required. This is enough to explore all working modes and run several real tasks. ## What affects credit usage Two main factors determine how many credits a task consumes: **1. The AI Mode selected** Every message runs in one of three modes. The mode sets the credit multiplier, and you can switch it per message right in the session composer: | AI Mode | Multiplier | Best for | | --------- | ---------- | ------------------------------------------------------------------------- | | **Lite** | 0.3Γ— | Light tasks at the lowest cost - scans, simple edits, routine content | | **Plus** | 1Γ— | The recommended default - balanced quality and speed for most tasks | | **Ultra** | 3Γ— | The strongest results - design work, strict content rules, complex builds | A quick fix can run on Lite while a page build in the same session runs on Ultra - the mode applies to the message, not the whole session. Lite runs on a mix of fast, lightweight models; Plus and Ultra use larger, more capable models. Plus and Ultra are available on paid plans. **2. The complexity of the task** A simple question like "how many posts do I have?" has a small cost. A complex task - building a plugin, generating a full page, running a multi-step agent - consumes significantly more credits because it involves multiple AI calls, web searches, code execution, and file operations. ## What real tasks cost Approximate costs from real runs, to give you a sense of scale: | Task | Credits | AI Mode | | --------------------------- | ------- | ------- | | Blog post with images | \~198 | Plus | | Broken links scan and fix | \~300 | Plus | | Full site SEO audit and fix | \~400 | Ultra | | Malware scan | \~86 | Lite | | Monthly site health report | \~132 | Plus | See the full breakdown on the [pricing page](https://foenix.ai/pricing). ## What else consumes credits * **Cloud Sites** are billed daily while active (12 to 30 credits/day depending on size) - see [Quick Start](/get-started/quick-start) for tiers. Your free 24-hour demo site costs nothing. * **Scheduled agents** consume credits on each run, at the same rates as running the task manually. ## Plans | Plan | Credits | Sites | | ----------------- | -------------- | ----------------- | | Free | 500 one-time | 1 demo site (24h) | | Starter - \$25/mo | 5,000 / month | 3 | | Pro - \$50/mo | 10,000 / month | 25 | Yearly billing saves 20%. Need more in a given month? On Starter and Pro, **credit packs** add 5,000 credits for \$19 without changing your plan. Full details on the [pricing page](https://foenix.ai/pricing). ## Tips to spend credits wisely Ask for a report before asking for changes. "Find the problems, report only" costs little and shows you the scope - then "fix everything from the report" does the work with no wasted correction loops. * **Be specific** - a clear, detailed instruction avoids back-and-forth correction loops that waste credits * **Pick the right mode** - not every task needs Ultra. Routine content and scans work well on Lite or Plus * **Test agents first** - run a task once manually before setting it as a recurring agent, so you know how much it costs in credits and how it performs * **Match the prompt to the job** - big builds like a full homepage work fine as one prompt. For everything else, smaller steps are cheaper, easier to control, and simpler to verify ## Checking your balance Your current credit balance is shown in the **bottom-left corner** of the Foenix dashboard. ## Running out of credits * **Free plan** - credits do not refresh. Upgrade to a paid plan to continue * **Starter / Pro** - credits refresh monthly. You can also purchase credit packs at any time without changing your plan # How It Works Source: https://docs.foenix.ai/core-concepts/how-it-works How Foenix connects to WordPress, the agent pipeline behind every task, the four autonomy levels, and the safety layers around them. Foenix connects to your WordPress site through a lightweight plugin and gives you a natural language interface to manage and edit your site. ## The two parts **Foenix Cloud** - the Foenix platform where AI agents receive your instructions in natural language, analyse the task, build a plan, and execute it using various tools - web search, browser access, code execution, and more. **Foenix Connector** - the plugin installed on your WordPress site. It is the bridge between your site and Foenix Cloud, communicating via the WP REST API. It reads your site content and applies changes. ## How a request works 1. You type a message in an open Foenix session 2. The **Orchestrator** analyses your request and your site's context, and routes the task 3. For heavy tasks the **Orchestrator** splits the work into parts and assigns each to a **Worker** sub-agent; the **Coder** executes step by step 4. The Connector deploys the changes on your WordPress site 5. The **Verifier** - a read-only agent - checks whether the task was actually completed, not just whether it ran. If verification fails, the task is retried with a different approach The agent that decides is never the agent that executes, and the agent that executes is never the one that grades the work. That separation is what makes autonomy safe. [Architecture β†’](/core-concepts/architecture) ## Four levels of autonomy From "nothing changes without me" to "runs while I sleep" - you pick the level per task: Ask questions about your site - structure, content, code. Read-only: Foenix only reads your site content for these requests, no changes are made. Click any element on your live site and describe the change. You point, the agent edits. Describe a full task and Foenix modifies your WordPress site - new content, edits, plugins, snippets, and more. Tasks that run automatically on a schedule - no chat open, no session running, no you. Foenix has full administrator-level access to your WordPress site and can modify files, database, plugins, and themes. Before running any edits or agents on a real site, make a backup, or rehearse the task in a Cloud Site first. [Backups β†’](/security/backups) ## Safety modes Foenix has 3 safety levels that control how strictly it validates actions before executing them. They work on top of the Verifier check that every change passes. In Soft mode the chance of something breaking is higher - always make a backup on your hosting before running any edits or agents. [Learn more β†’](/security/safety-mode) ## Data transfer security All communication between Foenix Cloud and the Connector is encrypted over **HTTPS**. Every request is authenticated - no unauthenticated access to your site is possible. If your firewall blocks incoming requests, switch the Connector to **pull-only mode** in its settings. The plugin then reaches out to Foenix Cloud itself, and no inbound connection to your server is needed. ## Abilities API Coming soon. If enabled in your account settings, Foenix will first look for Abilities API capabilities registered on your site and use them to complete the task. If none are found, it falls back to the standard WP REST API. # Sessions Source: https://docs.foenix.ai/core-concepts/sessions How sessions work in Foenix: starting, attaching context, AI Modes, stopping safely, history, and rollback. A session is a conversation thread between you and your WordPress site. Each session has a title, a history of messages, and a record of all actions taken. ## Starting a session 1. Open your Foenix dashboard and click a connected site 2. Click **New Session** in the left sidebar 3. Type your first message Each session is tied to a single site. Sessions are kept separate per site. ## What you can attach In any session you can attach additional context to your message: * **Images** - screenshots, mockups, reference photos * **PDF files** - documents, briefs, specifications * **Text files** - any plain text content * **ZIP archives** - upload a plugin or theme archive and Foenix unpacks it and reads the files inside * **YouTube links** - Foenix can read and use video content * **Any URL** - paste a link and Foenix will read the page ## Session duration Depending on the task complexity and the selected AI Mode, a session can complete in seconds or run for several minutes. Simple queries respond instantly. Complex tasks - like building a plugin or a theme - may take longer. You can close the browser tab while a session is running - it will continue working in the background, and you get an email when the task finishes. ## Stopping a session You can stop a session mid-run. The session keeps its context, so you can continue the conversation afterwards. Credits already spent are not refunded, and stopping mid-run may leave your site in an incomplete state - for example a plugin half-installed or a file partially modified. Check the result before continuing. ## One task per session We recommend creating a new session for each separate task. This keeps the context clean, reduces credit usage, and makes it easier to track what was done. ## Session history All past sessions appear in the **Recent Sessions** list in the left sidebar, sorted by most recent. Click any session to reopen it - continue the conversation or review what was done. ## Context within a session Foenix maintains context throughout a session. If you say "change the title of that post" after previously discussing a specific post, Foenix understands what you mean. Context does **not** carry over between chat sessions. Each new session starts fresh. Scheduled Agents are the exception: they keep a memory of your site between runs. [Scheduled Agents β†’](/working-modes/agents) ## AI Modes Every message runs in one of three modes - **Lite** (0.3Γ—), **Plus** (1Γ—, recommended), or **Ultra** (3Γ—). The mode sets the credit multiplier and the strength of the models behind it. Switch it per message right in the session composer: a quick fix can run on Lite while a page build in the same session runs on Ultra. Plus and Ultra are available on paid plans. [How modes affect cost β†’](/core-concepts/credits) Behind every message, Foenix runs a pipeline of AI agents - Orchestrator, Worker, Coder, and Verifier - regardless of the mode you pick. The mode changes the horsepower, not the process. [How the pipeline works β†’](/core-concepts/architecture) ## SERP settings If your task involves web search - for example writing an SEO article or researching competitors - Foenix uses the **SERP location and language** configured in **Settings - Personality**. Make sure these match your target market for best results. ## Rollback After actions that modify your site, a **Rollback** button may appear in the session. If something could not be undone, Foenix tells you right away instead of reporting a false success. [Learn more β†’](/security/rollback) ## If something goes wrong If you encounter errors or unexpected behaviour, reach out in our Discord community - the team responds quickly. [Join Discord β†’](https://discord.gg/Rammm3ts5u) # Use Cases Source: https://docs.foenix.ai/core-concepts/use-cases Real prompts for Foenix agents with approximate costs: content, SEO, site health, builds, and WooCommerce automation. Every example below is a complete, working prompt - paste it into a session as is, or adapt it to your site. Costs are approximate and depend on your site's size; the AI Mode listed is the recommended one for that task. [How credits and modes work β†’](/core-concepts/credits) Tasks marked **scheduled** include a schedule right in the prompt and keep running on their own. The rest are one-time runs you trigger when needed. ## Content & SEO ### Autonomous blog Β· scheduled A new article published every day: topic picked, text written, images generated, internal links added. ```text theme={null} Every day write and publish a blog post relevant to my website's niche, including 4 generated images: 1 featured image and 3 in-text body images. Auto-link to related past posts. ``` \~198 credits per post Β· Lite ### SEO audit and fix Β· scheduled Scans every post, fixes what can be fixed safely, and emails a report about the rest. ```text theme={null} Audit SEO on all posts. Find broken links, missing meta descriptions, missing alt texts. Fix everything and email me a report. ``` \~400 credits per run Β· Lite Β· works well weekly ### Make your site AI-ready Prepares your site for AI crawlers: creates `llms.txt`, checks `robots.txt` rules, and reports what was changed. ```text theme={null} Bring my site up to speed for AI crawlers: add an llms.txt with a summary of the site and links to the most important pages, check robots.txt for rules that block AI crawlers, and report what you changed. ``` Cost depends on site size Β· Lite ### Bulk content refresh Old posts updated in one run: dates, outdated stats, new internal links, regenerated excerpts. ```text theme={null} Update all posts older than 2 years: refresh dates, update outdated stats, add new internal links, regenerate excerpts. ``` \~1,200 credits Β· Lite ### Broken links auto-fix Β· scheduled Fixes what can be proven correct, quarantines the rest with suggestions. ```text theme={null} Scan all internal and external links weekly. Quarantine 404s, suggest replacements, fix automatically when confident. ``` \~300 credits per run Β· Lite ### Rank in Google Β· scheduled Weekly top-10 analysis for your keyword, applied to your page, with a summary of what changed and why. ```text theme={null} Every Wednesday, analyze the Google top-10 for my keyword, extract the SEO patterns they share, compare against my page, apply the improvements, and email me a summary. ``` \~350 credits per run Β· Lite ### SEO silo builder A complete topical cluster: hub article, supporting spokes, and the linking structure between them. ```text theme={null} Create Hub article (4000 words) + 5 supporting articles. Internal Silo linking: Hub↔Spokes↔siblings. ``` \~2,300 credits Β· Lite ### Premium data content A long-form hub with interactive charts built from real data, plus supporting articles. ```text theme={null} Run keyword research. Build a long-form Hub with embedded interactive charts from real data. 5 supporting articles, interlinked. ``` \~560 credits Β· Lite ## Site health ### Image optimization The whole media library compressed and converted, with pages verified to render identically. ```text theme={null} Compress all images in media library. Convert to WebP, add lazy loading, generate srcset. Target Lighthouse > 95. ``` \~2,800 credits Β· Lite ### Malware scan Every PHP file read in context, threats quarantined rather than deleted, exact paths reported. ```text theme={null} Scan all PHP files for base64_decode, eval, hidden iframes. Report with file paths. Quarantine confirmed threats. ``` \~86 credits Β· Lite Β· cheap enough to run daily ### Site health report Β· scheduled A monthly PDF about uptime, vulnerabilities, updates and performance, delivered to your inbox. ```text theme={null} Build a monthly site health report: uptime, vulnerabilities, updates, performance. Generate PDF and email it. ``` \~132 credits per run Β· Lite ## Build ### Instant homepage A rebuilt homepage tuned for speed, shipped as an isolated plugin so your theme stays untouched. ```text theme={null} Build me a new homepage that loads instantly. Target Lighthouse 100. ``` \~3,700 credits Β· Lite ## WooCommerce ### Store-wide update with translations Rebrand, reprice and translate an entire catalog in one pass. ```text theme={null} Replace 'OldBrand' with 'NewBrand' across all products, raise prices 18%, rewrite descriptions, translate DE/FR/ES. ``` \~1,200 credits Β· Lite ### Order alerts Β· event-triggered Β· coming soon Event-triggered agents are in development. When they ship, this prompt will fire on every new order - no schedule, no open chat, no session running: ```text theme={null} When a new WooCommerce order arrives, send instant Telegram notification with order details and total. ``` Before scheduling any of these, run it once manually. You'll see exactly what it costs on your site and how it performs - then add the schedule line to the same prompt. # What is Foenix.ai? Source: https://docs.foenix.ai/get-started/introduction AI agents that run your WordPress 24/7. Ask for something now, or set a task on a schedule: posts, pages, SEO fixes, site health. Your WordPress, on autopilot. Foenix is a team of AI agents that lives inside your WordPress: it understands your site's structure, builds anything you describe, and keeps working when you close the tab. Ask for something now, or set a task that runs on a schedule. Foenix writes posts, builds pages and plugins, fixes SEO, and keeps your site healthy while you sleep. Works with any WordPress: new or existing sites, any theme, any hosting. No migration needed.