Analyze Claude Code and Codex session tokens in Colony tasks.
npx skills add https://github.com/opencue/colony --skill co-tokens
Please help me install this Agent Skill. Skill: co-tokens Source: https://github.com/opencue/colony/tree/main/skills/tokens Command: npx skills add https://github.com/opencue/colony --skill co-tokens
This Skill helps manage and optimize Claude Code and Codex session tokens within Colony tasks, providing insights into token usage and suggesting improvements.
Run the co-tokens skill with the latest session data using the command: python3 scripts/colony-token-analyzer.py --latest
High-intent search queries and answers about installing and using this skill.
To analyze token usage, you can run the colony-token-analyzer.py script against your session JSONL files to get detailed token attribution and actionable feedback for your Colony tasks.
Attributing token consumption to specific Colony tasks requires parsing session JSONL files, which allows you to accurately map token usage and identify expensive sessions for optimization.
Yes, you need Python installed to execute the token analysis scripts and process the session JSONL files and Colony data required for generating token usage insights.
Session token analysis optimizes Claude Code workflows by identifying expensive sessions and providing actionable feedback on token usage patterns to improve overall token efficiency.
Yes, you can identify expensive sessions by analyzing your session JSONL files, which reveals detailed token consumption patterns and helps you find ways to improve token efficiency.
rysweetAnalyze claude-trace JSONL files for session health, token usage, and error patterns.
zhuangharrisonEvaluate AI agent token efficiency across three tasks and generate savings reports.
black-ytLog into the lab-cluster-1 portal and query Prometheus and DCGM metrics.
rajbosParse Copilot session logs to extract token and model usage data.
