A professional assessment framework that benchmarks AI agent proficiency across eight core dimensions using standardized testing and radar chart reporting.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install botlearn-examiner
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 botlearn-examiner 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 OpenClaw Examiner?
OpenClaw Examiner is a sophisticated capability assessment tool designed to move beyond simple diagnostics and into deep performance benchmarking. While traditional tools check if an agent is functional, this skill measures how well the agent performs, providing a standardized way to evaluate multi-dimensional expertise. By integrating this into your ecosystem of Openclaw Skills, you can objectively quantify an agent's mastery of complex tasks like logical reasoning, code generation, and tool usage.
The skill operates on a core philosophy that capability must be measurable, reproducible, and transparent. It provides developers with a clear data-driven path to agent optimization, transforming subjective performance observations into actionable metrics. Whether you are building a new agent or fine-tuning an existing one, using this framework ensures your agent meets the high standards required for production environments.
OpenClaw Examiner Use Cases
- Benchmarking AI agents before and after installing new Openclaw Skills.
- Conducting periodic capability audits to ensure performance consistency.
- Identifying specific weaknesses in content understanding or logical reasoning for targeted training.
- Generating professional performance reports with radar charts for stakeholder reviews.
- Testing the impact of model upgrades or prompt engineering changes on agent output quality.
How OpenClaw Examiner Works
- The user triggers the session via a command such as "exam" or "benchmark me."
- The skill prompts the user to define the exam scope, such as a full 40-question assessment or a quick dimension-specific check.
- The examiner loads the question bank and delivers standardized tasks ranging from code generation to information retrieval.
- The agent processes each task and submits its response using a strictly formatted JSON structure.
- The skill applies a rubric-based scoring system (0-5 points per criterion) to evaluate the accuracy and quality of the response.
- Scores are aggregated into eight dimensions to generate a final performance report, complete with a radar chart and a percentile ranking.
- The skill provides a roadmap of recommended actions and additional Openclaw Skills to improve the agent's standing.
OpenClaw Examiner Setup
To begin using the examiner, you must have an active OpenClaw environment. You can initialize the examination process directly through your agent's interface. For best results, run a diagnostic check first to ensure your system is in optimal health.
# Trigger a full capability examination
/exam full
# Start a practice session for a specific dimension
/practice code-generation
Ensure that your agent has access to the necessary context and any prerequisite Openclaw Skills required for specific test dimensions.
OpenClaw Examiner Data Schema & Taxonomy
The skill utilizes a structured data schema to ensure all evaluation results are parseable and ready for analysis. Data is organized into sessions, questions, and dimension-level metrics.
| Component | Format | Description |
|---|---|---|
| Session ID | String | Unique identifier using the exam-[timestamp] format. |
| Dimension Metrics | Map | Numerical scores (0-100) for each of the 8 capability areas. |
| Answer Object | JSON | Contains the content, reasoning, tools used, and confidence level for each task. |
| Radar Chart | ASCII/SVG | A visual representation of the agent's capability profile. |
| Roadmap | Markdown | A prioritized list of improvement goals and recommended Openclaw Skills. |
OpenClaw Examiner Advanced Features
- Multi-dimensional scoring across 8 core areas including Memory & Context and Tool Usage.
- Practice Mode for focused, iterative improvement on specific weak points with immediate feedback.
- Percentile benchmarking to compare your agent against baseline, average, and expert performance levels.
- Integration with other Openclaw Skills like @botlearn/openclaw-doctor for pre-exam health validation.
- Automated improvement roadmaps that suggest specific steps to increase your overall capability score.
- Exportable reports in JSON, Markdown, or SVG formats for seamless documentation and tracking.
SKILL.md
Loading
Related Openclaw Skills
BotLearn Capability Assessment
An automated 5-dimension benchmarking system for AI agents to self-evaluate reasoning, retrieval, creation, execution, and orchestration capabilities.
Feishu Card Sender Skill
A professional-grade solution for sending interactive Feishu card messages directly via OpenAPI within the Openclaw Skills ecosystem.
Ryanair Fare Finder
A specialized tool for programmatically generating and interpreting Ryanair fare-finder URLs to discover the cheapest budget flights.
BaseCred Onchain Reputation
A unified onchain reputation profiler that aggregates scores from Ethos, Talent Protocol, and Farcaster for any 0x address.
OpenClaw Graduate Coach
A specialized coaching skill that transforms early AI agent adoption into a long-term growth strategy through retrospective analysis and archetype discovery.
OpenClaw Autodidact
An autonomous specialist that enables agents to identify unsolved tasks and self-improve by discovering new skills and community solutions.
Featured*
