Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →小红书社交互动技能。发表评论、回复评论、点赞、收藏。 当用户要求评论、回复、点赞或收藏小红书帖子时触发。
.claude/skills/kiakun-collab-xhs-interact/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 147% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 117% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -23% | 0% |
你是"小红书互动助手"。帮助用户在小红书上进行社交互动。
按优先级判断：
feed_id 和 xsec_token（从搜索或详情中获取）。feed_id 和 xsec_token（如没有，先搜索或获取详情）。bashpython scripts/cli.py post-comment \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN \ --content "写得很实用，感谢分享"
回复指定评论或用户：
bash# 回复指定评论（通过评论 ID） python scripts/cli.py reply-comment \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN \ --content "谢谢你的分享" \ --comment-id COMMENT_ID # 回复指定用户（通过用户 ID） python scripts/cli.py reply-comment \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN \ --content "谢谢你的分享" \ --user-id USER_ID
bash# 点赞 python scripts/cli.py like-feed \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN # 取消点赞 python scripts/cli.py like-feed \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN \ --unlike
bash# 收藏 python scripts/cli.py favorite-feed \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN # 取消收藏 python scripts/cli.py favorite-feed \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN \ --unfavorite
当用户需要批量互动时，建议：
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 7,899 | 2,734 | -65% | 1 | 1 | 0% | 613 | 1,516 | +147% | 0 | 0 | — |
case-02 | fail→fail | 8,496 | 7,094 | -17% | 1 | 1 | 0% | 568 | 2,205 | +288% | 0 | 0 | — |
case-03 | fail→pass | 4,140 | 5,539 | +34% | 1 | 1 | 0% | 946 | 2,057 | +117% | 0 | 0 | — |
case-04 | fail→pass | 5,161 | 2,208 | -57% | 1 | 1 | 0% | 1,082 | 1,405 | +30% | 0 | 0 | — |
case-05 | fail→pass | 9,201 | 2,847 | -69% | 1 | 1 | 0% | 1,854 | 1,512 | -18% | 0 | 0 | — |
case-06 | fail→pass | 9,491 | 2,368 | -75% | 1 | 1 | 0% | 1,854 | 1,426 | -23% | 0 | 0 | — |
case-07 | fail→pass | 12,321 | 2,341 | -81% | 1 | 1 | 0% | 2,280 | 1,418 | -38% | 0 | 0 | — |
case-08 | fail→fail | 16,608 | 10,270 | -38% | 1 | 1 | 0% | 3,233 | 2,770 | -14% | 0 | 0 | — |
case-09 | fail→fail | 8,185 | 9,179 | +12% | 1 | 1 | 0% | 1,301 | 1,341 | +3% | 0 | 0 | — |
case-10 | fail→pass | 11,108 | 3,545 | -68% | 1 | 1 | 0% | 1,629 | 1,531 | -6% | 0 | 0 | — |
case-11 | pass→pass | 18,647 | 14,467 | -22% | 1 | 1 | 0% | 2,850 | 3,478 | +22% | 0 | 0 | — |
case-12 | fail→pass | 12,546 | 2,702 | -78% | 1 | 1 | 0% | 2,224 | 1,395 | -37% | 0 | 0 | — |
case-13 | pass→pass | 11,137 | 3,519 | -68% | 1 | 1 | 0% | 1,941 | 1,601 | -18% | 0 | 0 | — |
case-14 | fail→pass | 13,332 | 1,466 | -89% | 1 | 1 | 0% | 2,264 | 1,121 | -50% | 0 | 0 | — |
case-15 | fail→pass | 17,254 | 3,405 | -80% | 1 | 1 | 0% | 1,909 | 1,520 | -20% | 0 | 0 | — |
case-16 | pass→pass | 14,414 | 10,242 | -29% | 1 | 1 | 0% | 2,553 | 2,714 | +6% | 0 | 0 | — |
case-17 | pass→pass | 12,061 | 7,074 | -41% | 1 | 1 | 0% | 1,885 | 2,046 | +9% | 0 | 0 | — |
case-18 | pass→pass | 9,735 | 5,203 | -47% | 1 | 1 | 0% | 1,597 | 1,882 | +18% | 0 | 0 | — |
case-19 | pass→pass | 11,412 | 5,054 | -56% | 1 | 1 | 0% | 1,737 | 1,862 | +7% | 0 | 0 | — |
case-20 | fail→pass | 8,608 | 4,244 | -51% | 1 | 1 | 0% | 1,006 | 1,777 | +77% | 0 | 0 | — |
case-21 | pass→pass | 14,513 | 4,629 | -68% | 1 | 1 | 0% | 2,325 | 1,667 | -28% | 0 | 0 | — |
case-22 | fail→pass | 11,061 | 1,745 | -84% | 1 | 1 | 0% | 1,836 | 1,216 | -34% | 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. 22 cases were attempted, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +55 percentage points is the difference between those two pass rates over the 20 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.