skills.rest
rbuke
rbukeCOMMUNITY
Explore all skills by this author →
GitHub

performance-reviewer

Grade autonomous AI agents on quality, actionability, completeness, and impact.

Updated Mar 9, 2026
Search Tags:#feedback loop#ai evaluation#agent management#performance review#agent grading#ceo assistant
One-click install
npx skills add https://github.com/rbuke/project-jabroni --skill performance-reviewer-rbuke
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: performance-reviewer
Source: https://github.com/rbuke/project-jabroni/tree/main/.claude/skills/performance-reviewer
Command: npx skills add https://github.com/rbuke/project-jabroni --skill performance-reviewer-rbuke

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3, and includes scripts (resource) components.

What problem does it solve?

This Skill automates the evaluation of AI agent performance, providing objective feedback and identifying areas for improvement or potential termination.

Core Features & Use Cases

  • Automated Grading: Assigns scores to agents based on quality, actionability, completeness, and impact of their output.
  • Prescriptive Feedback: Offers specific, actionable advice for underperforming agents.
  • Escalation: Flags agents for potential termination to the CEO based on consistent poor performance.
  • Use Case: After a development cycle, the performance-reviewer agent analyzes the code commits, design documents, and user feedback generated by other agents, providing a clear scorecard and improvement plan for each.

Quick Start

Run the performance reviewer skill to grade all agent outputs since the last review.

Frequently Asked Questions about performance-reviewer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I evaluate autonomous AI agent performance and grade their output quality?▼

To evaluate autonomous AI agent performance, this skill assigns grades based on quality, actionability, completeness, and impact. It reads agent logs via filesystem communication to generate objective scorecards and prescriptive feedback for each agent.

What is the best way to provide automated feedback to underperforming AI agents?▼

Automated feedback for underperforming AI agents is generated by evaluating their actionability and completeness against predefined criteria. The skill delivers prescriptive, actionable advice directly to a shared board to drive immediate improvement.

How do I flag an autonomous agent for termination to a CEO agent?▼

Flagging an agent for termination to a CEO agent happens automatically when consistent poor performance is detected. The skill escalates underperformers by posting review summaries to a shared filesystem-based communication board.

Can I use this performance review skill with python3 scripts for agent management?▼

Yes, you can use this performance review skill with python3, as it explicitly requires python3 as a dependency. The skill utilizes scripts to analyze agent logs and post evaluation summaries to a shared board.

How does filesystem-based communication work for AI agent grading?▼

Filesystem-based communication for AI agent grading works by reading agent logs directly from the file system and posting review summaries to a shared board. This approach allows autonomous agents to exchange performance data without direct API integrations.

When do I need an automated agent grading system for my AI workflow?▼

You need an automated agent grading system when managing multiple autonomous agents and requiring objective evaluation of their output quality and impact. It is essential for identifying underperformers and driving continuous improvement after development cycles.

Related Skills

Explore More Product & Management Skills→
Atrium-HermesAtrium-Hermes

Evening Recap

Synthesize end-of-day operational summaries from memory and log data.

Community
Intermediate
nateanatea
21

Processing Email Inbox

Automate Gmail inbox processing with GTD triage and task creation.

Community
Advanced
blakesimsblakesims

orchestrate

Orchestrate multi-agent task workflows across planning, execution, and review phases.

Community
Advanced
abhishekmmgnabhishekmmgn

agent-quality-flywheel

Implement a four-step flywheel converting production data into agent improvements.

Community
Advanced