Agent Skills › WingedGuardian/GENesis-AGI › triage-calibration

triage-calibration

GitHub

用于每日或特定事件后校准分类准确性。通过对比近期决策与实际结果，计算准确率及误判率，识别系统性偏差，并提出权重或阈值调整建议以优化分类模型。

src/genesis/skills/triage-calibration/SKILL.md WingedGuardian/GENesis-AGI

Trigger Scenarios

每日定时校准发现重大分类错误时新增或修改信号采集器后

Install

npx skills add WingedGuardian/GENesis-AGI --skill triage-calibration -g -y

More Options

Non-standard path

npx skills add https://github.com/WingedGuardian/GENesis-AGI/tree/main/src/genesis/skills/triage-calibration -g -y

Use without installing

npx skills use WingedGuardian/GENesis-AGI@triage-calibration

指定 Agent (Claude Code)

npx skills add WingedGuardian/GENesis-AGI --skill triage-calibration -a claude-code -g -y

安装 repo 全部 skill

npx skills add WingedGuardian/GENesis-AGI --all -g -y

预览 repo 内 skill

npx skills add WingedGuardian/GENesis-AGI --list

SKILL.md

Frontmatter

{
    "name": "triage-calibration",
    "phase": 6,
    "consumer": "daily_calibration",
    "skill_type": "hybrid",
    "description": "Daily triage accuracy calibration — use during scheduled calibration runs to verify triage classification accuracy against few-shot examples and adjust confidence thresholds"
}

Triage Calibration

Purpose

Review recent triage decisions against actual outcomes. Adjust signal weights and depth thresholds to improve classification accuracy over time.

When to Use

Scheduled daily (end of day or low-activity period).
After a significant misclassification is identified.
After a new signal collector is added or modified.

Workflow

Collect recent triage results — Pull the last 24h of awareness ticks with their depth classifications and signal readings.
Match against outcomes — For each tick, determine what actually happened:
- Did the classified depth match the actual work required?
- Were any signals misleadingly high or low?
Compute accuracy metrics —
- Classification accuracy (correct depth / total ticks)
- Over-triage rate (classified higher than needed)
- Under-triage rate (classified lower than needed)
Identify systematic errors — Are certain signal types consistently miscalibrated? Are certain time periods problematic?
Propose adjustments — Suggest weight or threshold changes. Do NOT auto-apply in V3 (fixed weights). Log recommendations for user review.
Write calibration report.

Output Format

date: <YYYY-MM-DD>
period: <start> to <end>
total_ticks: <n>
accuracy: <percentage>
over_triage_rate: <percentage>
under_triage_rate: <percentage>
systematic_errors:
  - signal: <signal name>
    bias: over | under
    magnitude: <how far off>
recommendations:
  - <proposed adjustment>