skillsbench

🥈Silver

SkillsBench evaluates skill performance and agent effectiveness. Operations teams use it to optimize workflows. It connects to Claude agents and PDDL language.

9281720Updated 3d ago

Intermediate30min to implementautomation

Saves ~135 min per use

Quick InstallView Source

git clone https://github.com/benchflow-ai/skillsbench.git

Works with:

Claude

Overview

About This Skill

SkillsBench evaluates skill performance and agent effectiveness. Operations teams use it to optimize workflows. It connects to Claude agents and PDDL language.

How to Use

["1. **Define the Scope:** Clearly specify the agent name, workflow, and time period you want to evaluate. This will help the AI focus on the relevant data and metrics.","2. **Analyze the Output:** Review the AI's evaluation to understand the agent's strengths and weaknesses. Pay attention to the specific metrics provided for each area.","3. **Implement Improvements:** Use the suggested actionable improvements to optimize the agent's performance. This could involve training, process changes, or tool enhancements.","4. **Monitor Progress:** After implementing improvements, use SkillsBench to monitor the agent's performance over time to ensure the changes are effective.","5. **Iterate:** Regularly evaluate the agent's performance and make iterative improvements to continuously optimize workflows."]

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/benchflow-ai/skillsbench

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Evaluate the performance of the [AGENT_NAME] agent in handling [WORKFLOW_NAME] workflows over the past [TIME_PERIOD]. Identify the top 3 areas where the agent is excelling and the top 3 areas needing improvement. Provide specific metrics and suggest actionable improvements for each area of improvement.

Example Output

After evaluating the performance of the 'Customer Support Agent' in handling 'Ticket Resolution' workflows over the past month, here are the key findings:

**Top 3 Areas of Excellence:**
1. **Response Time:** The agent consistently resolves 85% of tickets within the SLA of 4 hours, with an average response time of 1.5 hours.
2. **Customer Satisfaction:** The agent maintains a customer satisfaction score of 92%, which is 15% higher than the team average.
3. **Knowledge Base Utilization:** The agent effectively uses the knowledge base to resolve 70% of tickets without escalation.

**Top 3 Areas for Improvement:**
1. **Escalation Rate:** The agent escalates 20% of tickets, which is higher than the team average of 12%. Suggested improvement: Implement a pre-escalation checklist to ensure all possible resolution steps are attempted.
2. **Follow-up Response Time:** The agent's follow-up response time averages 6 hours, which is above the SLA. Suggested improvement: Set up automated reminders for follow-up responses.
3. **Ticket Categorization Accuracy:** The agent miscategorizes 15% of tickets. Suggested improvement: Provide additional training on ticket categorization guidelines and best practices.