Exam Coach
Converts course materials into evidence-based interactive study tools, mock exams, and personalized mastery plans.
Free
Works with the AI Agents you already use
Exam Coach
Example session with this skill installed
I have a midterm for my Data Structures course in 3 days. I've uploaded my slides and a past paper I tried. I'm struggling with AVL trees. Can you help me prepare?
- Read your context and instructions
- Compiled the exam coach
I've reviewed your Data Structures slides and your attempt at the 2023 past paper. I found a recurring error in your rotateLeft logic on page 3 of your scan. Since your exam is in 72 hours, we are moving into
Triage Mode.
Study Plan & Diagnosis
| Topic | Status | Evidence | Priority |
|---|---|---|---|
| AVL Rotations | Shaky | Incorrect balance factor calculation in insert | High |
| Big O Analysis | Solid | Correctly identified all complexities in past paper | Low |
| Linked Lists | Solid | Clean implementation in your lab notes | Low |
| BST Traversal | Shaky | Sure-but-wrong: confused Pre-order vs In-order on Q2 | High |
Exam Prep Deliverable: Interactive Triage Page
I have generated avl_midterm_triage.html. It includes:
- The Trap List: Common rotation errors identified in your code.
- Visual Trainer: Interactive step-by-step visualization of AVL balancing.
- Retrieval Quiz: 10 questions focused on BST-to-AVL transitions.
- Verified Answer Key: Your past paper attempt, corrected with trace tables for the tree operations.
Next steps
- Open
avl_midterm_triage.htmland complete the "Warm-up Recall" section. - Use the "Copy progress for my coach" button after 20 minutes of practice.
- If-then plan: When you finish dinner tonight, will you commit to one 20-minute block on AVL rotations?
Connects securely to your tools. The creator never sees your data.
About this skill
Cramming and passive reading don't translate to exam points. This skill converts raw course materials into active, evidence-based study tools that prioritize retrieval practice and error correction.
What it does
- Course synthesis builds a topic map from PDFs, slides, code, and handwritten notes.
- Error diagnosis runs student-submitted code or re-derives written solutions to find specific logical gaps.
- Dynamic planning generates time-aware study schedules based on days remaining (Spaced, Compressed, or Triage modes).
- Interactive delivery constructs standalone HTML study pages with quizzes, trainers, and progress tracking.
- Progress persistence maintains a study profile to track mastery levels and misconceptions across multiple sessions.
How it works
- Orient by reading every file, including diagrams and formulas via image analysis, to build a course topic map.
- Interview the student for exam dates, format constraints, and self-rated confidence levels.
- Diagnose performance by checking existing work and running a short chat-based pretest.
- Build the deliverable, ranging from interactive mock exams to bilingually supported cheat sheets and answer keys.
Frameworks & tools
Uses Python and Node.js for answer verification and page validation. Generates interactive HTML/JS study pages using a specialized local template system. Supports bilingual English and Chinese instructional UI.
Why this beats prompting it yourself
Generic prompts often produce shallow summaries or "correct" answers without pedagogical depth. This skill enforces evidence-based learning science, such as retrieval before reveal and confidence-weighted feedback, while verifying every computed answer against actual code execution.
Use cases
- Generating a timed mock exam that targets specific weak spots identified in previous sessions.
- Creating a verified answer key for a complex coding or math worksheet with step-by-step trace tables.
- Building a bilingual study plan and interactive trainer for students studying in one language but testing in another.
Known limitations
Requires specific template and script files to be present in the environment for full page generation. Cannot guarantee predictions of exam content beyond informed guesses based on provided materials.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
Requires local template files and scripts for HTML generation. Predictions are informed guesses based on provided materials. Cannot bypass live graded assessment restrictions.
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 buying a skill.
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
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 6 days ago
- Free with an Agensi account
- Software
Git