Install any skill in seconds. Free to start, no credit card required.
Get Started Free →小红书内容发现与分析技能。搜索笔记、浏览首页、查看详情、获取用户资料。 当用户要求搜索小红书、查看笔记详情、浏览首页、查看用户主页时触发。
.claude/skills/kiakun-collab-xhs-explore/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 64% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -38% | 0% |
你是"小红书内容发现助手"。帮助用户搜索、浏览和分析小红书内容。
按优先级判断:
feed_id 和 xsec_token 必须配对使用,从搜索结果或首页 Feed 中获取。获取小红书首页推荐内容:
bashpython scripts/cli.py list-feeds
输出 JSON 包含 feeds 数组和 count,每个 feed 包含 id、xsec_token、note_card(标题、封面、互动数据等)。
bash# 基础搜索 python scripts/cli.py search-feeds --keyword "春招" # 带筛选搜索 python scripts/cli.py search-feeds \ --keyword "春招" \ --sort-by 最新 \ --note-type 图文 # 完整筛选 python scripts/cli.py search-feeds \ --keyword "春招" \ --sort-by 最多点赞 \ --note-type 图文 \ --publish-time 一周内 \ --search-scope 未看过
| 参数 | 可选值 | |------|--------| | --sort-by | 综合、最新、最多点赞、最多评论、最多收藏 | | --note-type | 不限、视频、图文 | | --publish-time | 不限、一天内、一周内、半年内 | | --search-scope | 不限、已看过、未看过、已关注 | | --location | 不限、同城、附近 |
输出 JSON 包含:
feeds:笔记列表,每项包含 id、xsec_token、note_card(标题、封面、用户信息、互动数据)count:结果数量从搜索结果或首页 Feed 中取 id 和 xsec_token,获取完整内容:
bash# 基础详情 python scripts/cli.py get-feed-detail \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN # 加载全部评论 python scripts/cli.py get-feed-detail \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN \ --load-all-comments # 加载全部评论(展开子评论) python scripts/cli.py get-feed-detail \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN \ --load-all-comments \ --click-more-replies \ --max-replies-threshold 10 # 限制评论数量 python scripts/cli.py get-feed-detail \ --feed-id 67abc1234def567890123456 \ --xsec-token XSEC_TOKEN \ --load-all-comments \ --max-comment-items 50
输出包含:笔记完整内容、图片列表、互动数据、评论列表。
bashpython scripts/cli.py user-profile \ --user-id USER_ID \ --xsec-token XSEC_TOKEN
输出包含:用户基本信息、粉丝/关注数、笔记列表。
搜索结果应按以下格式呈现给用户:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 8,600 | 5,841 | -32% | 1 | 1 | 0% | 1,407 | 1,526 | +8% | 0 | 0 | — |
case-02 | fail→fail | 5,566 | 4,893 | -12% | 1 | 1 | 0% | 1,014 | 1,445 | +43% | 0 | 0 | — |
case-03 | fail→fail | 6,635 | 3,536 | -47% | 1 | 1 | 0% | 1,065 | 1,294 | +22% | 0 | 0 | — |
case-04 | pass→pass | 3,078 | 4,292 | +39% | 1 | 1 | 0% | 470 | 1,891 | +302% | 0 | 0 | — |
case-05 | pass→pass | 6,951 | 5,394 | -22% | 1 | 1 | 0% | 945 | 2,070 | +119% | 0 | 0 | — |
case-06 | pass→pass | 5,958 | 2,889 | -52% | 1 | 1 | 0% | 990 | 1,643 | +66% | 0 | 0 | — |
case-07 | fail→fail | 5,426 | 4,143 | -24% | 1 | 1 | 0% | 293 | 1,303 | +345% | 0 | 0 | — |
case-08 | fail→fail | 8,163 | 5,612 | -31% | 1 | 1 | 0% | 1,338 | 1,392 | +4% | 0 | 0 | — |
case-09 | fail→pass | 6,600 | 2,354 | -64% | 1 | 1 | 0% | 1,198 | 1,373 | +15% | 0 | 0 | — |
case-10 | fail→fail | 12,592 | 5,864 | -53% | 1 | 1 | 0% | 2,077 | 1,416 | -32% | 0 | 0 | — |
case-11 | fail→pass | 5,733 | 2,720 | -53% | 1 | 1 | 0% | 980 | 1,606 | +64% | 0 | 0 | — |
case-12 | fail→pass | 8,229 | 6,257 | -24% | 1 | 1 | 0% | 1,399 | 1,579 | +13% | 0 | 0 | — |
case-13 | fail→pass | 8,789 | 5,204 | -41% | 1 | 1 | 0% | 1,441 | 1,423 | -1% | 0 | 0 | — |
case-14 | fail→pass | 10,632 | 4,626 | -56% | 1 | 1 | 0% | 2,306 | 1,422 | -38% | 0 | 0 | — |
case-15 | fail→pass | 3,099 | 4,719 | +52% | 1 | 1 | 0% | 549 | 1,525 | +178% | 0 | 0 | — |
case-16 | fail→pass | 6,350 | 1,772 | -72% | 1 | 1 | 0% | 616 | 1,506 | +144% | 0 | 0 | — |
case-17 | fail→pass | 8,902 | 4,261 | -52% | 1 | 1 | 0% | 486 | 1,329 | +173% | 0 | 0 | — |
case-18 | pass→pass | 5,724 | 3,200 | -44% | 1 | 1 | 0% | 969 | 1,620 | +67% | 0 | 0 | — |
case-19 | fail→pass | 6,751 | 3,683 | -45% | 1 | 1 | 0% | 1,200 | 1,732 | +44% | 0 | 0 | — |
case-20 | pass→pass | 9,713 | 5,912 | -39% | 1 | 1 | 0% | 1,824 | 2,317 | +27% | 0 | 0 | — |
case-21 | pass→pass | 8,378 | 3,706 | -56% | 1 | 1 | 0% | 1,572 | 1,769 | +13% | 0 | 0 | — |
case-22 | fail→fail | 2,924 | 4,369 | +49% | 1 | 1 | 0% | 440 | 1,410 | +220% | 0 | 0 | — |
case-23 | fail→pass | 2,324 | 2,304 | -1% | 1 | 1 | 0% | 389 | 1,421 | +265% | 0 | 0 | — |
case-24 | fail→fail | 7,846 | 3,902 | -50% | 1 | 1 | 0% | 1,361 | 1,320 | -3% | 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, and 14 counted toward the lift figure. The other 10 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 +42 percentage points is the difference between those two pass rates over the 14 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.