AI tutor chatbot for your course site
SleekAI reads your course content, lesson progress, and assessment rubric from WordPress, then explains concepts at the student’s level, walks through worked examples, and surfaces the next exercise. Plug in OpenAI, Anthropic, Google, or OpenRouter with your own key.
Static course content can’t answer the question a student actually has
A pre-recorded lesson and a quiz at the end of the module work fine when the material lands cleanly. The moment a student misunderstands one prerequisite, the rest of the module falls apart.
Forum posts get answered in two days, office hours run once a week, and the student fakes their way through the next quiz hoping it doesn’t matter. By the time the instructor notices the gap, the student has either dropped the course or built three more misconceptions on top of the first one.
The bot explains a concept at the student’s stated level, walks through a worked example using the same notation the lesson used, and surfaces the next exercise when the student is ready. Display conditions scope the tutor to the course area only, so the same bot doesn’t show up on the public marketing pages.
The recurring confusions reveal the lesson that needs a rewrite, the worked example that’s missing, and the prerequisite nobody links to. That feedback loop is the curriculum research artefact instructors usually only get from end-of-course surveys.
Features
Course-grounded
Every explanation references the lesson it comes from and uses the same notation and terminology the course already established. No reinventing the vocabulary mid-stream, which is what most generic AI tutors do.
Student-aware
The explanation lands at the right depth, references prerequisites by name, and doesn’t pretend the student already knows something they haven’t covered.
Answers from your data
Map post titles, content, post meta, taxonomies or a WP_Query into the instruction, and the bot answers from what your site actually publishes.
Where instructors use SleekAI as a tutor
Cohort-based courses
Live cohort programs where students need help between live sessions.
Self-paced courses
Asynchronous programs where students stall on a single concept and disappear. The bot intervenes at the moment of friction with a worked example, which is the difference between completion and refund.
Skill bootcamps
Coding, design, or trades programs with rubric-graded exercises. The bot uses the rubric to explain what “better” looks like on a draft submission, without doing the work for the student.
How SleekAI runs as a one-on-one tutor
Ground in the course
ACF and postmeta fields on lesson posts carry prerequisite links and notation so the bot uses the course’s vocabulary.
Map your data
Choose the posts, fields, taxonomies or users the bot should know and reference them in the instruction.
Set display conditions
Decide which pages, post types, roles or logged-in states show the chatbot.
Mine the gaps
The recurring confusions become the basis for a clarifying note, a new worked example, or a transcript rewrite that lifts the next cohort’s outcomes.
Why grounded tutoring beats a generic AI helper
Most courses lose students at a single concept that didn’t land. Forum posts get answered in two days, office hours run once a week, and the student fakes their way through the next quiz hoping it doesn’t matter.
By the time the instructor notices the gap, the student has either dropped the course or built three more misconceptions on top of the first one. A grounded tutor closes that gap by intervening at the moment of friction, in the course’s own notation, using lessons the student has actually completed as the grounding for the explanation. The operational win is completion.
Students who get a grounded explanation at the moment of confusion finish at a higher rate, because the friction that would have ended the course gets resolved in minutes instead of days. The pedagogical win is depth.
A tutor that uses the course’s notation reinforces the vocabulary the curriculum spent two lessons establishing, rather than introducing a competing standard from a textbook the student isn’t reading. The longer-term win is curriculum research. The right metric is not chats handled but lesson completion within the cohort’s expected timeline, which is the input the course’s reputation and renewal rate actually respond to.
Frequently asked questions
Will the bot do the student’s homework for them?
Only if you let it. The system prompt should explicitly limit the bot to explaining concepts, walking through example problems that are not the assigned exercises, and pointing the student to the rubric. Most instructors instruct the bot to refuse to answer the current assignment directly and to offer a worked example on a parallel problem instead.
Does it use the course’s notation?
If the lesson transcripts and worked examples are in the grounding context, the bot picks up the notation from those sources. The system prompt should also reinforce “use the notation from the course, not a textbook standard you might know”, which keeps the bot from quietly switching variable names mid-explanation.
Can it recommend the next exercise?
Yes. The grounding can include the exercise index per lesson, and the prompt instructs the bot to recommend an exercise that builds on the concept the student just asked about. If the LMS exposes a prerequisite graph, the bot follows it; if not, the prompt encodes the recommended order explicitly.
How does it handle different student levels in the same course?
Include the student’s self-rated level or a placement-quiz outcome in the user meta, and the prompt adjusts depth accordingly. An advanced student gets a tighter explanation that assumes prerequisites; a beginner gets a worked example and a pointer to the prerequisite lesson. Same course, different entry points.
What about regulated topics like medical or legal coursework?
The bot should never give medical, legal, or financial advice as if it were a practitioner. The course’s own disclaimers should be referenced in the prompt so the bot can quote them when needed.
Can it grade rubric-based exercises?
The grading judgement stays with the instructor for accountability reasons; the bot’s job is to coach the student to a better draft before submission, which is where most rubric-based feedback adds the most value.
Which AI providers does it support?
OpenAI, Anthropic, Google and OpenRouter, plus custom OpenAI-compatible endpoints. You bring your own API key and the provider bills you directly.
Where does the chatbot get its answers?
From the WordPress data you map into its instruction: post titles and content, post meta, ACF and Meta Box fields, taxonomies, terms, users and user meta, or a WP_Query. Values resolve on every request.
Pricing
Explore our flexible licensing options tailored to your needs. Upgrade your license anytime to access more features, or opt for a lifetime license for ongoing value, including lifetime updates and lifetime support. Our hassle-free upgrade process ensures that our platform can grow with you, starting from whichever plan you choose.
…or get the Bundle Deal
and save €250
The Bundle (unlimited sites)
Pay once, own it forever
Elevate your WordPress site with our exclusive plugin bundle that includes all of our premium plugins in one package. Enjoy lifetime updates and lifetime support. Save significantly compared to buying plugins individually.
What’s included
SleekAI
SleekByte
SleekMotion
SleekPixel
SleekView
SleekRank
SleekTrack