How accurate are the token estimates provided by this skill?

Token estimates are approximate heuristics based on conversation length and message counts, designed to provide a general overview of resource consumption rather than exact billing data.

How do I trigger the dashboard report?

You can view the full session dashboard at any time by using the /metrics command within your Claude Code environment.

Why do some metrics display as 'N/A'?

Metrics may show 'N/A' if the required conversation-level tracking data is unavailable to a specific sub-agent or if the session history is too short to provide a statistical sample.

What is the 'Parallelism Efficiency' metric?

This metric calculates the ratio of parallel batches to total batches observed in the session, helping you understand how effectively the system is utilizing concurrent task execution.

Does the Metrics skill require write permissions?

No, this is a read-only diagnostic skill that analyzes existing session data and does not modify your codebase or project files.

Session Performance Metrics

Name: Session Performance Metrics
Author: eroslifestyle

byeroslifestyle

0•

Analytics & Monitoring

Monitors and displays a comprehensive dashboard of agent performance, task completion rates, and parallelism efficiency for Claude orchestration sessions.

The Metrics skill provides deep observability into Claude Code sessions by aggregating real-time data on agent activity and task execution. It generates a detailed dashboard that tracks task status, ranks agent utilization, estimates token consumption, and calculates parallelism efficiency. This tool is essential for developers using multi-agent systems who need to audit performance, identify bottlenecks in concurrent workflows, or monitor error recovery rates during complex coding tasks.

Key Features

01Session duration and estimated token consumption monitoring

02Parallelism efficiency analysis and batch processing statistics

03Top 5 agent usage ranking by task invocation count

040 GitHub stars

05Comprehensive task status tracking (Completed, Failed, In Progress, Pending)

06Error tracking with breakdown of auto-recovered versus manual interventions

Use Cases

01Generating a post-session summary to track learning metrics and system reliability

02Auditing agent efficiency during complex multi-agent software development sessions

03Optimizing workflow parallelism to reduce session duration and resource usage

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add eroslifestyle/claude-orchestrator-plugin metrics

For use in Claude.ai and ChatGPT

Key Features

01Session duration and estimated token consumption monitoring

02Parallelism efficiency analysis and batch processing statistics

03Top 5 agent usage ranking by task invocation count

040 GitHub stars

05Comprehensive task status tracking (Completed, Failed, In Progress, Pending)

06Error tracking with breakdown of auto-recovered versus manual interventions

Use Cases

01Generating a post-session summary to track learning metrics and system reliability

02Auditing agent efficiency during complex multi-agent software development sessions

03Optimizing workflow parallelism to reduce session duration and resource usage

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add eroslifestyle/claude-orchestrator-plugin metrics

For use in Claude.ai and ChatGPT