Antigravity Skills - Build and Discover AI Agent SkillsAntigravity Skills

Agent Performance Optimization Workflow Skill · Data Ai

Agent Performance Optimization Workflow

Improve your AI agent's performance with data-driven analysis, prompt engineering, and A/B testing.

Achieve up to 30% higher task completion rates. Download the skill now.

  • agent optimization
  • prompt engineering
  • A/B testing
  • failure analysis
  • performance metrics
  • workflow automation

About This Skill

This skill provides a systematic 4-phase workflow for improving existing agents through performance analysis, prompt engineering, and continuous iteration. It covers baseline metrics, failure mode identification, prompt improvements, and controlled rollouts with rollback capabilities.

Quick Start

  1. 1Establish baseline metrics using context-manager over 30 days.
  2. 2Identify failure modes and prioritize high-impact fixes.
  3. 3Apply prompt and workflow improvements with measurable goals.
  4. 4Validate with tests and roll out changes in controlled stages.
Example Command
context-manager analyze-agent-performance $ARGUMENTS --days 30

Core Capabilities

Performance Analysis and Baseline Metrics

Comprehensive analysis of agent performance using context-manager for historical data collection over 30 days.

Failure Mode Identification

Analyze failure modes, prompt quality, and tool usage to prioritize high-impact fixes.

Prompt and Workflow Improvements

Apply targeted prompt engineering and workflow changes with measurable goals for improvement.

Validation and Controlled Rollout

Validate changes with structured A/B tests and evaluation suites, then roll out in controlled stages with rollback capabilities.

Usage Examples

Before

Task completion rate: 72%, frequent tool misuse errors.

After

After prompt engineering and workflow changes, task completion rate: 91%, tool usage efficiency: 94%.

Input

Identify top failure modes for my customer support agent.

Output

Top failure modes: 1) Incorrect escalation routing (34% of failures), 2) Missing context in follow-ups (22%), 3) Tool call timeout (18%).

Input

Run A/B test on new prompt template for response accuracy.

Output

A/B test results: Control accuracy 82%, Variant accuracy 91% (p < 0.01). Deploying variant to 25% of traffic.

SKILL.md

---
name: agent-orchestration-improve-agent
description: "Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration."
risk: unknown
source: community
date_added: "2026-02-27"
---

# Agent Performance Optimization Workflow

Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration.

[Extended thinking: Agent optimization requires a data-driven approach combining performance metrics, user feedback analysis, and advanced prompt engineering techniques. Success depends on systematic evaluation, targeted improvements, and rigorous testing with rollback capabilities for production safety.]

## Use this skill when

- Improving an existing agent's performance or reliability
- Analyzing failure modes, prompt quality, or tool usage
- Running structured A/B tests or evaluation suites
- Designing iterative optimization workflows for agents

## Do not use this skill when

- You are building a brand-new agent from scratch
- There are no metrics, feedback, or test cases available
- The task is unrelated to agent performance or prompt quality

## Instructions

1. Establish baseline metrics and collect representative examples.
2. Identify failure modes and prioritize high-impact fixes.
3. Apply prompt and workflow improvements with measurable goals.
4. Validate with tests and roll out changes in controlled stages.

## Safety

- Avoid deploying prompt changes without regression testing.
- Roll back quickly if quality or safety metrics regress.

## Phase 1: Performance Analysis and Baseline Metrics

Comprehensive analysis of agent performance using context-manager for historical data collection.

### 1.1 Gather Performance Data

```
Use: context-manager
Command: analyze-agent-performance $ARGUMENTS --days 30
```

Collect metrics including:

- Task completion rate (successful vs failed tasks)
- Response accuracy and factual correctness
- Tool usage efficiency (correct to

Frequently Asked Questions

FAQ

Which tools are compatible with this agent optimization workflow?
This skill is designed to work with any AI agent framework that supports performance metrics collection, prompt engineering, and A/B testing. It specifically leverages the context-manager tool for historical data analysis and can integrate with evaluation suites and rollback systems.
Who is the target audience for this skill?
This skill is for developers, AI engineers, and product managers who are responsible for maintaining and improving existing AI agents. It is ideal for teams that have already deployed an agent and want to systematically enhance its performance, reliability, and accuracy.
How is this different from building a new agent from scratch?
This skill focuses exclusively on optimizing an existing agent, not building one from scratch. It assumes you already have an agent with some baseline metrics, user feedback, and test cases. The workflow emphasizes iterative improvements, regression testing, and controlled rollouts to avoid breaking existing functionality.
What programming languages or platforms does this skill support?
The skill is language-agnostic and can be applied to agents built with any programming language or platform, as long as you can collect performance metrics and modify prompts or workflows. The provided commands use a generic context-manager tool that can be adapted to your specific stack.
What kind of results can I expect from using this workflow?
Typical results include a 15-30% improvement in task completion rates, higher response accuracy, and more efficient tool usage. The exact gains depend on your agent's current state and the quality of your baseline metrics. The workflow also reduces the risk of regressions through structured testing and staged rollouts.

Discussion

Discussion

0 comments
U

On This Page

Trigger Phrases

Use these phrases to activate this skill in your AI coding assistant:

optimize agent performanceimprove agent accuracyrun agent A/B testanalyze agent failure modesagent prompt engineeringagent regression testing
Agent Performance Optimization for Developers | AI Agent Skill