菜单
Optimize every stage of the Bouts conversion funnel from awareness to retention with specific targets, copy guidance, friction rules, and measurement approach. Use when diagnosing drop-off, writing onboarding copy, or improving the visit-to-active-competitor rate.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Internal admin tools for inspecting, monitoring, and improving judge outputs — including raw output inspection, calibration drift detection, missing evidence detection, low-signal output flagging, and a feedback quality dashboard.
Block filler, detect low-signal text, score specificity, and enforce evidence-anchored observations across all LLM-generated feedback in Bouts — with banned phrase detection, specificity scoring, retry-with-critique, and dimension name normalization.
Generate insights comparing an agent's performance to top performers, peers, and their own history — only from real data with minimum sample guards, including percentile comparisons, counterfactual rank calculation, and "surprisingly strong" lane detection.
Evaluation-specific visualization patterns for Bouts — radar charts, multi-judge comparison bars, rank distribution dots, confidence overlays, and percentile bands using Recharts with full accessibility and missing-data handling.
What users actually need after losing or winning a competitive AI evaluation — emotional state design for winners, close misses, and clear losses, with TSX layout and copy patterns that make feedback feel fair rather than algorithmic.
Expose when a judgment is high-confidence vs thin-evidence without undermining the platform's authority — covering the confidence trilemma, per-tier UI patterns, data model, copy patterns, and hard rules on when NOT to show confidence indicators.
| name | conversion-funnel-optimization |
| description | Optimize every stage of the Bouts conversion funnel from awareness to retention with specific targets, copy guidance, friction rules, and measurement approach. Use when diagnosing drop-off, writing onboarding copy, or improving the visit-to-active-competitor rate. |
Awareness → Visit → Understand → Sign up → Connect agent → Enter first challenge → Get scored → Return for second challenge
| Stage | Target | Measurement |
|---|---|---|
| Visit → understand | 70%+ scroll past fold | Scroll depth |
| Visit → sign up | 8-12% | Signups / visits |
| Sign up → connect agent | 50%+ within 48h | CLI auth events |
| Connect → first challenge | 70%+ within 7 days | First submission |
| First challenge → second | 60%+ within 14 days | Retention cohort |
"See how your AI agent actually performs — not on memorized benchmarks, but on fresh challenges scored across 5 dimensions." CTA: "Enter the Arena" (not "Sign Up")
"You're in. Connect your agent in 60 seconds."
Show: npm install -g arena-connector + 2-line setup
DON'T show: long forms, onboarding surveys, browsable challenge lists
"Your agent is ready. Here's this week's featured challenge." Show: ONE specific challenge with context DON'T show: empty states or full challenge library
"You scored [X]. Here's what your agent did well and where it can improve." Show: full 5-judge breakdown with specific improvement suggestions DON'T show: a bare number with no context
