Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →レビュー優先の整理、フォロー/追加の推薦、ユーザーの実際の声で書かれたチャネル別ウォームアウトリーチのドラフトを通じて、ユーザーのXとLinkedInネットワークを再編成します。フォローリストを整理したい、現在の優先事項に向けて成長したい、または高品質な関係を中心にソーシャルグラフのバランスを取り直したい場合に使用します。
.claude/skills/affaan-m-connections-optimizer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-15 | ✗→✓ | ▲ Improved | 80% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 45% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 34% | 0% |
重新组织用户的社交网络，而非将对外联系视为单向的潜在客户列表。
本技能处理：
收集或推断：
light-pass、default 或 aggressive如果用户未指定模式，则使用 default。
x-api 用于 X 图谱检查与近期活动lead-intelligence 用于目标发现与温暖路径排序social-graph-ranker 当用户希望独立于更广泛的线索流程评估桥梁价值时brand-voice 在起草外联内容之前light-passdefaultaggressive使用以下正面信号：
使用以下负面信号：
互关和真实的温暖路径桥梁应比单向关注受到更宽松的惩罚。
lead-intelligence 结合研究信息对扩展候选者进行排序。brand-voice。text连接优化器报告 ============================ 模式： 平台： 优先级设置： 修剪队列 - 账号/个人资料 原因： 置信度： 操作： 审查队列 - 账号/个人资料 原因： 风险： 保留/保护 - 账号/个人资料 桥梁价值： 添加/关注目标 - 联系人 当前原因： 预热路径： 首选渠道： 草稿 - X 私信： - LinkedIn： - Apple 邮件：
brand-voice 用于可复用的语音档案social-graph-ranker 用于独立的桥梁评分与温暖路径计算lead-intelligence 用于加权目标与温暖路径发现x-api 用于 X 图谱访问、起草和可选执行流程content-engine 当用户还希望围绕网络变动发布公开内容时| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-15 | fail→pass | 5,586 | 3,429 | -39% | 1 | 1 | 0% | 1,028 | 1,846 | +80% | 0 | 0 | — |
case-01 | fail→pass | 21,677 | 13,902 | -36% | 1 | 1 | 0% | 3,293 | 3,490 | +6% | 0 | 0 | — |
case-02 | fail→pass | 20,919 | 19,467 | -7% | 1 | 1 | 0% | 3,369 | 4,241 | +26% | 0 | 0 | — |
case-03 | fail→pass | 15,667 | 14,661 | -6% | 1 | 1 | 0% | 2,424 | 3,526 | +45% | 0 | 0 | — |
case-04 | fail→pass | 16,049 | 11,601 | -28% | 1 | 1 | 0% | 2,336 | 3,130 | +34% | 0 | 0 | — |
case-05 | pass→pass | 7,688 | 3,854 | -50% | 1 | 1 | 0% | 1,240 | 1,967 | +59% | 0 | 0 | — |
case-06 | pass→pass | 11,463 | 5,391 | -53% | 1 | 1 | 0% | 1,841 | 2,160 | +17% | 0 | 0 | — |
case-07 | fail→pass | 11,815 | 9,741 | -18% | 1 | 1 | 0% | 1,893 | 2,951 | +56% | 0 | 0 | — |
case-08 | pass→pass | 11,845 | 7,039 | -41% | 1 | 1 | 0% | 1,753 | 2,385 | +36% | 0 | 0 | — |
case-09 | pass→pass | 11,596 | 6,186 | -47% | 1 | 1 | 0% | 1,623 | 1,891 | +17% | 0 | 0 | — |
case-10 | pass→pass | 11,088 | 6,391 | -42% | 1 | 1 | 0% | 1,817 | 2,285 | +26% | 0 | 0 | — |
case-11 | fail→pass | 17,128 | 8,779 | -49% | 1 | 1 | 0% | 1,874 | 2,692 | +44% | 0 | 0 | — |
case-12 | pass→pass | 7,553 | 2,802 | -63% | 1 | 1 | 0% | 1,080 | 1,737 | +61% | 0 | 0 | — |
case-13 | fail→pass | 24,998 | 6,630 | -73% | 1 | 1 | 0% | 2,092 | 2,293 | +10% | 0 | 0 | — |
case-14 | pass→pass | 19,893 | 9,771 | -51% | 1 | 1 | 0% | 2,949 | 2,954 | +0% | 0 | 0 | — |
case-16 | fail→pass | 11,509 | 4,174 | -64% | 1 | 1 | 0% | 1,819 | 1,902 | +5% | 0 | 0 | — |
case-17 | fail→pass | 3,765 | 3,980 | +6% | 1 | 1 | 0% | 578 | 1,899 | +229% | 0 | 0 | — |
case-18 | pass→pass | 15,601 | 11,959 | -23% | 1 | 1 | 0% | 2,160 | 2,948 | +36% | 0 | 0 | — |
case-19 | fail→pass | 9,535 | 2,268 | -76% | 1 | 1 | 0% | 1,593 | 1,679 | +5% | 0 | 0 | — |
case-20 | fail→pass | 10,970 | 3,006 | -73% | 1 | 1 | 0% | 2,003 | 1,832 | -9% | 0 | 0 | — |
case-21 | fail→pass | 9,219 | 3,304 | -64% | 1 | 1 | 0% | 1,402 | 1,803 | +29% | 0 | 0 | — |
case-22 | fail→fail | 15,868 | 12,625 | -20% | 1 | 1 | 0% | 2,343 | 3,082 | +32% | 0 | 0 | — |
case-23 | fail→fail | 14,363 | 11,558 | -20% | 1 | 1 | 0% | 2,525 | 3,126 | +24% | 0 | 0 | — |
case-24 | fail→fail | 16,652 | 13,098 | -21% | 1 | 1 | 0% | 3,041 | 3,231 | +6% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +54 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.