An advanced multimodal AI diagnostic skill designed to analyze exam papers, classify student errors, and generate adaptive variation exercises into standalone, print-friendly HTML reports.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install exam-paper-error-analysis
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install exam-paper-error-analysis using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Exam Paper Error Analysis?
The Exam Paper Error Analysis skill is a highly sophisticated agentic workflow built for educators, curriculum developers, and academic coaches. Operating as a key asset among Openclaw Skills, this tool automates deep diagnostics of student exam papers across primary, secondary, and vocational education. By accepting direct text inputs or analyzing uploaded images and multi-page PDFs, the skill converts raw test documents into detailed, interactive reports without relying on heavy external OCR APIs.
At its core, the skill uses multimodal AI reasoning to pinpoint exactly why a student missed a point, differentiating conceptual misunderstandings from procedural slips. For vocational education subjects—ranging from CNC machining to nursing—it dynamically maps student errors directly to national vocational standards and workplace competencies. This makes it an invaluable addition to any AI-assisted grading setup using Openclaw Skills.
Exam Paper Error Analysis Use Cases
- Single-Question Deep Diagnostics: Educators can paste or upload a single incorrect answer to generate a multi-step behavioral and cognitive breakdown of why the student failed.
- Whole-Paper Score Loss Mapping: Grade an entire student paper (via photos or PDFs) to produce a structured, visual HTML map of error categories and cumulative score losses.
- Automated Variation Question Generation: Instantly generate targeted practice exercises (consolidation, common trap, and transfer questions) tailored precisely to a student's observed weaknesses.
- Class-Wide Remediation Planning: Classify group-level mistakes across six distinct cognitive error types to construct a prioritized classroom feedback and lecturing guide.
How Exam Paper Error Analysis Works
- Intent Recognition and Routing: The agent scans user inputs for specific keywords (such as error analysis, test paper mapping, or generate practice) to automatically trigger one of the four core sub-capabilities.
- Multimodal Input Processing: Standard text files are parsed instantly, while PDF and image uploads are routed through a secure, local pipeline. PDFs are rendered page-by-page into PIL images at scale=2 using pypdfium2 to optimize multimodal recognition.
- Multimodal AI Analysis: The native vision capabilities of the AI model process the rendered images to extract questions, student answers, teacher marks, and scratchpad work directly without external OCR API calls.
- Cognitive Error Classification: The skill assesses the mistake complexity level (Basic, Intermediate, Advanced) and aligns it against six standardized error archetypes detailed in reference modules.
- Standardized HTML Output Generation: The diagnostic results, interactive index navigation, custom-generated practice sheets, and lecturing advice are synthesized into a single print-friendly HTML document.
Exam Paper Error Analysis Setup
To run the Exam Paper Error Analysis skill within your agent workspace, ensure you have the required Python library pypdfium2 installed for local PDF-to-image processing.
Install the dependencies:
pip install pypdfium2 Pillow
Ensure that your workspace contains the required reference templates in the references/ directory:
references/html-template.md: Defines the structured HTML/CSS export schema.references/error-types.md: Provides error taxonomy rules.references/subject-adaptation.md: Custom rules for K12 and vocational subjects.references/vocational-standards.md: Mapping matrices for professional skills.
Once files are in place, invoke your AI agent utilizing Openclaw Skills, upload a test paper (image or PDF), and issue an analysis prompt.
Exam Paper Error Analysis Data Schema & Taxonomy
The skill processes inputs dynamically and compiles all findings into a highly structured, self-contained HTML file. Below is the mapping of data properties generated during the pipeline:
| Chapter Section | Field | Purpose |
|---|---|---|
| 1. Basic Information | Grade, Subject, Paper Name, Total Points |
Provides context and metadata categorization for the diagnostic report |
| 2. Navigation Index | Table of Contents | Employs inline anchor tags allowing seamless navigation between report sections |
| 3. Full-Paper Loss Analysis | Error Stats, Question Types, Mapping Chart | Aggregates score deficits and groups them by cognitive error categories |
| 4. Question-by-Question Analysis | Question Text, Error Classification, Correction | Breaks down individual answers using a strict seven-step diagnostic framework |
| 5. Adaptive Practice Generation | Consolidation, Common Trap, and Transfer Questions | Generates custom variation exercises with step-by-step model keys |
| 6. Lecture and Review Guide | Lecturing Priority, Teaching Strategies | Outlines direct, actionable pedagogical advice for classroom teachers |
All HTML reports are written directly to disk using the naming format: 错题分析报告_Grade_Subject_Date.html.
Exam Paper Error Analysis Advanced Features
- Multi-Page PDF Processing Pipeline: Includes an integrated, scriptless PDF rendering pipeline that uses pypdfium2 to parse extensive documents page-by-page, strictly enforcing a page-limit safety alert for files exceeding 20 pages.
- Vocational Standards Alignment: Dynamically switches on vocational adaptation modes when analyzing professional subjects like CNC machining, electrical wiring, or nursing, auto-flagging critical safety or protocol violations as high-risk.
- Evidence-Driven AI Reasoning: Enforces strict cognitive boundaries ensuring that no assumptions are made about a student's overall capabilities; every diagnostic claim must be backed by visible work-scratch evidence.
- Print-Ready HTML Generation: Embeds responsive CSS media queries specifically optimized for standard print layouts, allowing teachers to print physical handouts directly from the browser using Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Vocational Curriculum Standard Generator (curriculum-standard)
An advanced document generation tool that automates the translation of vocational training plans and professional teaching guidelines into MoE-compliant DOCX syllabus files.
Table Specification (Exam Blueprint Generator)
An AI coding agent skill designed to systematically translate educational syllabi into highly structured, scientifically balanced test item specification tables and exam blueprints.
LYGO Champion Council
A consolidated multi-agent advisor skill providing access to all 15 Δ9 quantum council champion personas with a strict advisor-only execution policy.
Simple Todo Manager
A lightweight, no-frills todo list manager for capturing, tracking, and completing daily tasks within your workspace.
Score Analyzer
An automated, local data-analysis pipeline that cleans, tags, and processes student score sheets from Excel, producing professional diagnostic reports and rich visual charts without needing external API dependencies.
Fidacy Conversation Receipts
A secure developer tool that creates local, tamper-evident cryptographic receipts for AI agent conversations to prevent legal disputes and verify bot statements.
Featured*
