Back to Details

skillsmp

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

SkillsMP

SkillsMP

搜 SkillsMP —— 目前最大的公开 Agent Skill 索引（1.6M+ 个 SKILL.md，来自 GitHub，覆盖 Claude Code、Codex、ChatGPT）。
要动手写一个新 Skill 之前，先来这里搜一遍。 别人写过的概率比你以为的高。
Search SkillsMP —— the largest public Agent Skill index currently available (1.6M+ SKILL.md files from GitHub, covering Claude Code, Codex, ChatGPT).
Before writing a new Skill, search here first. The probability that someone has already written it is higher than you think.

目录

Table of Contents

skillsmp/
├── SKILL.md            ← 你在这里：怎么搜、怎么挖宝、两个必须知道的坑
├── .env.example        ← 复制成 .env 填 API Key（可选；.env 已被忽略，绝不提交）
└── scripts/
    ├── search.mjs      直搜。可翻页、可按分类/职业/语言过滤，可 --json
    └── treasure.mjs    ★ 挖宝。故意不按星数排，理由见下
skillsmp/
├── SKILL.md            ← You are here: how to search, how to find hidden gems, two critical pitfalls to know
├── .env.example        ← Copy to .env and fill in API Key (optional; .env is ignored and will never be committed)
└── scripts/
    ├── search.mjs      Direct search. Supports pagination, filtering by category/occupation/language, and --json
    └── treasure.mjs    ★ Treasure hunting. Intentionally not sorted by stars, see below for reasons

先跑起来

Get Started

不需要任何配置——匿名就能用（50 次/天、10 次/分钟）：
bash
node scripts/search.mjs "关键词" --limit 20
想要 500 次/天，就配一个 Key：
bash
cp .env.example .env      # 然后把 Key 填进去
Key 从 https://skillsmp.com/docs/api 生成。它是凭据，只放 
.env
，绝不进仓库 （仓库的 
.gitignore
已经拦了 
*/.env
，别绕过它）。
No configuration required — use anonymously (50 requests/day, 10 requests/minute):
bash
node scripts/search.mjs "关键词" --limit 20
For 500 requests/day, configure a Key:
bash
cp .env.example .env      # Then fill in your Key
Generate the Key from https://skillsmp.com/docs/api. It is a credential, only store it in 
.env
and never commit it to the repository (the repository's 
.gitignore
already blocks 
*/.env
, do not bypass this).

两个必须知道的坑

Two Critical Pitfalls to Know

★ 不是这个 Skill 的星数，是它所在仓库的星数

★ The star count is for the repository, not the Skill

这是本 Skill 存在的主要理由。API 返回的 
stars
是包含该 Skill 的 GitHub 仓库 的星数。实测：某条结果报 240467，而它所在仓库的真实星数是 240743 —— 对得上，确认无疑。
后果很实际：
  • 一个塞在超高星仓库里的 Skill（哪怕只是整包机翻的文档）自动继承那个星数；
  • 一个作者单独开仓库、认真写的单一用途 Skill，只有个位数星。
实测搜 
backlink
按星排序，前四条里三条来自同一个 28562★ 的笔记仓库， 讲的是笔记系统内部的双向链接，跟外链毫无关系。高星把语义对口的结果整个淹掉了。
所以：
sortBy=stars
排出来的不是「最好的 Skill」，是「住在最红仓库里的 Skill」。
This is the main reason this tool exists. The 
stars
returned by the API is the star count of the GitHub repository containing the Skill. Verification: A result showed 240467 stars, while the actual star count of its repository was 240743 — they match, confirming this fact.
Practical consequences:
  • A Skill embedded in a high-star repository (even if it's just machine-translated documentation) automatically inherits that star count;
  • A single-purpose Skill carefully written by an author in an independent repository may only have single-digit stars.
When searching for 
backlink
sorted by stars, three of the top four results come from the same 28562★ note repository, discussing internal bidirectional links in a note system, which has nothing to do with external links. High-star results completely drown out semantically relevant ones.
Therefore: Sorting by 
sortBy=stars
does not give you the "best Skills", but rather "Skills residing in the most popular repositories".

翻页要认 
hasNext
，别认 
total

Use 
hasNext
for pagination, not 
total

pagination.total
附带一个 
totalIsExact: false
，而且实测严重偏低——百万级索引里 搜 
SEO
只报 
total: 5
。按 total 算页数会漏掉绝大部分结果。 唯一可靠的翻页依据是 
hasNext
，两个脚本都已经这么做了。
pagination.total
comes with 
totalIsExact: false
, and in practice it is severely underestimated — searching for 
SEO
in a million-level index only returns 
total: 5
. Calculating pages based on total will miss most results. The only reliable pagination indicator is 
hasNext
, which both scripts already use.

挖宝：找「写得好但没人知道」的

Treasure Hunting: Find "Well-Written but Unknown" Skills

既然星数不是质量信号，就别用它排。
treasure.mjs
用四个跟仓库名气无关的信号：
bash
node scripts/treasure.mjs "关键词" --pages 5 --max-stars 2000 --top 15
信号想法
独立性所在仓库星数越低，越说明这个 Skill 靠自己站住，不是搭便车
专注度同一仓库在本次结果里出现几条。一个仓库刷出几十条，通常是批量生成或整包翻译的文档堆
描述具体度好的描述会写清什么时候该用（触发条件、场景、反例），而不是「帮你做 X」。这是分辨用不用心最单一有效的信号
新鲜度长期没动的多半已经烂掉
同名同作者跨多语言的条目会被去重——那是整包机翻，一个仓库能刷满整页。
这是启发式排序，不是判决。 脚本只负责把候选排到你眼前； 要不要用，仍然得打开那个 
SKILL.md
读一遍。别把分数当成质量结论报给用户。
Since star count is not a quality signal, don't use it for sorting. 
treasure.mjs
uses four signals unrelated to repository popularity:
bash
node scripts/treasure.mjs "关键词" --pages 5 --max-stars 2000 --top 15
SignalRationale
IndependenceThe lower the star count of the repository it resides in, the more it indicates this Skill stands on its own, not riding on the coattails of a popular repo
FocusHow many times the same repository appears in the results. A repository showing up dozens of times usually means bulk-generated or machine-translated documentation dumps
Specificity of DescriptionGood descriptions clearly state when to use (trigger conditions, scenarios, counterexamples), rather than just "helps you do X". This is the single most effective signal to distinguish thoughtful Skills
FreshnessSkills that haven't been updated in a long time are likely outdated
Entries with the same name and author across multiple languages are deduplicated — these are bulk machine-translated entries that can fill an entire page from one repository.
This is heuristic sorting, not a judgment. The script only ranks candidates for you to review; you still need to open the 
SKILL.md
and read it to decide whether to use it. Don't treat the score as a definitive quality conclusion for users.

过滤参数

Filter Parameters

两个脚本共用：
参数说明
--pages N
翻几页（search 默认 1，treasure 默认 5）
--limit N
每页几条，上限 100
--sort stars|recent
只有 search 有。先读上面那条坑再决定用不用 stars
--category <slug>
如 
data-ai
、
devops
--occupation <slug>
SOC 职业，如 
software-developers
--lang <code>
en
/ 
zh
/ 
ja
等 ISO 码；
mul
混合，
und
判不出
--json
输出 JSON 而不是表格
不支持通配符（
*
），也不支持空查询。
Shared by both scripts:
ParameterDescription
--pages N
Number of pages to paginate (default 1 for search, 5 for treasure)
--limit N
Number of results per page, maximum 100
--sort stars|recent
Only available for search. Read the pitfall above before deciding whether to use stars
--category <slug>
e.g., 
data-ai
, 
devops
--occupation <slug>
SOC occupation, e.g., 
software-developers
--lang <code>
ISO codes like 
en
/ 
zh
/ 
ja
; 
mul
for multilingual, 
und
for undetermined
--json
Output JSON instead of table
Wildcards (
*
) and empty queries are not supported.

配额与报错

Quotas and Error Handling

响应头一直在报剩余量，脚本会把它打在结尾。常见错误已经翻译成人话： 
INVALID_API_KEY
（Key 无效）、
DAILY_QUOTA_EXCEEDED
（当日用完）、 
MISSING_QUERY
（没给关键词）、
INVALID_OCCUPATION
/ 
INVALID_LANGUAGE
（slug 不认识）。
搜索结果尽量不要在一次任务里反复重搜同一个词——配额是按天算的， 匿名只有 50 次。需要反复查询时用 
--json
存一份到本地再过滤。
The response header always shows remaining quota, which the script will print at the end. Common errors have been translated into plain language: 
INVALID_API_KEY
(invalid Key), 
DAILY_QUOTA_EXCEEDED
(daily quota used up), 
MISSING_QUERY
(no keyword provided), 
INVALID_OCCUPATION
/ 
INVALID_LANGUAGE
(unknown slug).
Avoid repeatedly searching the same term in one task — quotas are daily, and anonymous users only get 50 requests. For repeated queries, use 
--json
to save results locally and filter offline.

汇报纪律

Reporting Guidelines

把搜到的 Skill 告诉用户时：
  • 说清 ★ 是仓库星数，不要让用户以为那是这个 Skill 的受欢迎程度；
  • 推荐之前至少读一眼它的 
    githubUrl
    ，别只凭 description 就推荐—— 描述是作者自己写的营销文案，不是验证过的能力；
  • 命中很少时如实说命中很少。这个索引有 1.6M 条，搜不到通常意味着词不对， 换个说法再搜一次，而不是断言「不存在」。
When sharing found Skills with users:
  • Clearly state that ★ refers to the repository's star count, don't let users think it reflects the Skill's popularity;
  • At least check the 
    githubUrl
    before recommending     — don't recommend based solely on the description, since descriptions are self-written marketing copy, not verified capabilities;
  • Be honest when there are few results. With 1.6M entries in the index, not finding results usually means the keyword is incorrect, so try rephrasing instead of claiming "none exist".