Use when tracking team performance, analyzing agent strengths and weaknesses, collecting metrics, and generating performance reports. Trigger with performance review or metrics requests.
Use when tracking team performance, analyzing agent strengths and weaknesses, collecting metrics, and generating performance reports. Trigger with performance review or metrics requests.
user-invocable
false
license
Apache-2.0
compatibility
Requires AI Maestro for agent metrics, session memory for historical data, and reporting capabilities. Requires AI Maestro installed.
metadata
{"author":"Emasoft","version":"1.0.0"}
context
fork
agent
ecos-main
workflow-instruction
support
procedure
support-skill
Emasoft Chief of Staff Performance Tracking Skill
Overview
Performance tracking enables the Chief of Staff to understand how well the agent team is performing, identify individual strengths and weaknesses, and make data-driven decisions about team composition and task assignment. This skill teaches systematic approaches to collecting metrics, analyzing performance, and generating useful reports.
Prerequisites
Before using this skill, ensure:
Agent metrics are being collected
Performance baselines are established
Reporting templates are available
Instructions
Identify performance metrics needed
Query agent activity logs
Calculate performance indicators
Generate performance report
Output
Metric Type
Output
Task completion
Completion rate, average time
Resource usage
Memory, CPU, API calls
Quality metrics
Error rate, rework rate
What Is Performance Tracking?
Performance tracking is the systematic collection and analysis of metrics about agent behavior, output quality, and efficiency. Unlike traditional employee performance reviews, agent performance tracking happens continuously and focuses on measurable outcomes rather than subjective assessments.
Key characteristics:
Data-driven: Based on measurable metrics, not impressions
Continuous: Ongoing tracking, not point-in-time reviews
Actionable: Insights lead to concrete improvements
Fair: Consistent criteria applied to all agents
Performance Tracking Components
1. Performance Metrics
Quantifiable measures of agent output and behavior.
2. Strength-Weakness Analysis
Identification of what each agent does well and where they struggle.
3. Performance Reporting
Communication of findings to relevant stakeholders.
Core Procedures
PROCEDURE 1: Collect Performance Metrics
When to use: Continuously during agent operation, at task completion, during periodic reviews.
Steps: Define metrics to track, capture data at relevant events, aggregate over time periods, validate data quality.
# Agent Analysis: helper-agent-generic## Strengths1.**Code Review Speed**: Completes reviews 25% faster than average
2.**First-Pass Quality**: 90% of code passes review on first attempt
3.**Communication**: Clear, concise status updates
## Weaknesses1.**Complex Algorithms**: Struggles with optimization tasks
2.**Documentation**: Often leaves docs incomplete
3.**Estimation**: Underestimates by average of 30%
## Recommendations- Assign code review tasks (strength)
- Pair with senior agent for algorithm work (weakness mitigation)
- Require documentation checklist for all tasks
Example 3: Weekly Performance Summary
# Weekly Performance Summary
Period: 2025-01-27 to 2025-02-02
## Team Overview- Active Agents: 8
- Tasks Completed: 45
- On-Time Rate: 82%
- First-Pass Quality: 75%
## Top Performers1. libs-svg-svgbbox: 12 tasks, 100% on-time
2. helper-agent-generic: 8 tasks, 88% first-pass
## Areas for Improvement1. Documentation completion rate: 65% (target: 90%)
2. Estimation accuracy: +35% average overrun
## Recommendations- Add documentation checkpoint to workflow
- Review estimation process with underperforming agents
Operational Procedures
Step-by-step runbooks for executing each performance tracking operation. Use these when performing the actual procedures described above.
