Skip to content

Agent skills study 2026: what 10,000 popular skills do

10,000 agent skills from 2,841 public GitHub repositories with 100 to 10,000 stars, sorted by the job each does and the field it serves, with the AI products they name, how they are built and their licences.

0:00
13:42
skills examined
10,000
repositories with 100 to 10,000 stars
2,841
made for a field outside technology
34%
of repositories with no licence detected
12%

Main findings

  1. Writing and changing code is the most common job, at 1,314 (13.1%) of the 10,000 skills, followed by building and running AI agents at 10.9%.
  2. 33.5% of the skills are made for a field outside technology; Science and research leads with 828 (8.3%).
  3. Popularity differs little by job: the median repository holding each kind of skill has between 280 and 397 stars. By field, health and sales skills sit in the most-starred repositories (437 and 435).
  4. Claude is named in 21.8% of the skills, Codex in 7.3% and Gemini in 3.2%; 73.2% name no AI product at all.
  5. 353 (12.4%) of the 2,841 repositories carry no licence GitHub can detect, so 14.3% of the skills come with no stated permission to reuse them.

Every figure describes the 10,000 skills examined on 20 November 2026, from repositories with 100 to 10,000 stars. Repositories with more than 10,000 stars are reported apart.

What the skills do

Writing and changing code is the most common job of popular skills, at 13%Write and change code: 1,314 (13.1%). Build and run AI agents: 1,087 (10.9%). Work with data: 880 (8.8%). Plan and manage work: 777 (7.8%). Research and gather information: 759 (7.6%). Deploy and run systems: 685 (6.8%). Test and fix bugs: 656 (6.6%). Check security: 638 (6.4%). Run business tasks: 569 (5.7%). Write and edit text: 474 (4.7%). Market and sell: 429 (4.3%). Design interfaces: 428 (4.3%). Review code: 422 (4.2%). Make images, video and audio: 352 (3.5%). Make documents and slides: 220 (2.2%). Other: 156 (1.6%). Teach and explain: 154 (1.5%)Write and change code1,314 (13.1%)Build and run AI agents1,087 (10.9%)Work with data880 (8.8%)Plan and manage work777 (7.8%)Research and gather information759 (7.6%)Deploy and run systems685 (6.8%)Test and fix bugs656 (6.6%)Check security638 (6.4%)Run business tasks569 (5.7%)Write and edit text474 (4.7%)Market and sell429 (4.3%)Design interfaces428 (4.3%)Review code422 (4.2%)Make images, video and audio352 (3.5%)Make documents and slides220 (2.2%)Other156 (1.6%)Teach and explain154 (1.5%)Writing and changing code is the most common job of popular skills, at 13%Write and change code: 1,314 (13.1%). Build and run AI agents: 1,087 (10.9%). Work with data: 880 (8.8%). Plan and manage work: 777 (7.8%). Research and gather information: 759 (7.6%). Deploy and run systems: 685 (6.8%). Test and fix bugs: 656 (6.6%). Check security: 638 (6.4%). Run business tasks: 569 (5.7%). Write and edit text: 474 (4.7%). Market and sell: 429 (4.3%). Design interfaces: 428 (4.3%). Review code: 422 (4.2%). Make images, video and audio: 352 (3.5%). Make documents and slides: 220 (2.2%). Other: 156 (1.6%). Teach and explain: 154 (1.5%)Write and change code1,314 (13.1%)Build and run AI agents1,087 (10.9%)Work with data880 (8.8%)Plan and manage work777 (7.8%)Research and gather information759 (7.6%)Deploy and run systems685 (6.8%)Test and fix bugs656 (6.6%)Check security638 (6.4%)Run business tasks569 (5.7%)Write and edit text474 (4.7%)Market and sell429 (4.3%)Design interfaces428 (4.3%)Review code422 (4.2%)Make images, video and audio352 (3.5%)Make documents and slides220 (2.2%)Other156 (1.6%)Teach and explain154 (1.5%)
Base: 10,000 skills from 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026. One main job for each skill.
Main job, with the repositories that hold such skills and their median stars (base: 10,000 skills)
Main jobSkillsShareRepositoriesMedian stars
Write and change code1,31413.1%610349
Build and run AI agents1,08710.9%637308
Work with data8808.8%372287
Plan and manage work7777.8%469291
Research and gather information7597.6%368343
Deploy and run systems6856.8%431356
Test and fix bugs6566.6%436334
Check security6386.4%229309
Run business tasks5695.7%219328
Write and edit text4744.7%326342
Market and sell4294.3%189348
Design interfaces4284.3%281307
Review code4224.2%317345
Make images, video and audio3523.5%239328
Make documents and slides2202.2%158397
Other1561.6%93284
Teach and explain1541.5%100280

Which fields they serve

60% of popular skills are made for technology work, 34% for other fieldsSoftware development: 2,467 (24.7%). AI and data: 1,132 (11.3%). Web and mobile apps: 955 (9.6%). Science and research: 828 (8.3%). No particular field: 688 (6.9%). Marketing: 493 (4.9%). Cybersecurity: 491 (4.9%). Cloud and IT: 440 (4.4%). Design and creative work: 377 (3.8%). Finance: 357 (3.6%). Personal productivity: 249 (2.5%). Legal: 224 (2.2%). Media and publishing: 209 (2.1%). Games: 197 (2.0%). E-commerce: 183 (1.8%). Hardware and engineering: 179 (1.8%). Sales and customer service: 129 (1.3%). Education: 123 (1.2%). Health: 110 (1.1%). Blockchain and crypto: 101 (1.0%). HR and recruiting: 49 (0.5%). Real estate: 19 (0.2%)Software development2,467 (24.7%)AI and data1,132 (11.3%)Web and mobile apps955 (9.6%)Science and research828 (8.3%)No particular field688 (6.9%)Marketing493 (4.9%)Cybersecurity491 (4.9%)Cloud and IT440 (4.4%)Design and creative work377 (3.8%)Finance357 (3.6%)Personal productivity249 (2.5%)Legal224 (2.2%)Media and publishing209 (2.1%)Games197 (2.0%)E-commerce183 (1.8%)Hardware and engineering179 (1.8%)Sales and customer service129 (1.3%)Education123 (1.2%)Health110 (1.1%)Blockchain and crypto101 (1.0%)HR and recruiting49 (0.5%)Real estate19 (0.2%)60% of popular skills are made for technology work, 34% for other fieldsSoftware development: 2,467 (24.7%). AI and data: 1,132 (11.3%). Web and mobile apps: 955 (9.6%). Science and research: 828 (8.3%). No particular field: 688 (6.9%). Marketing: 493 (4.9%). Cybersecurity: 491 (4.9%). Cloud and IT: 440 (4.4%). Design and creative work: 377 (3.8%). Finance: 357 (3.6%). Personal productivity: 249 (2.5%). Legal: 224 (2.2%). Media and publishing: 209 (2.1%). Games: 197 (2.0%). E-commerce: 183 (1.8%). Hardware and engineering: 179 (1.8%). Sales and customer service: 129 (1.3%). Education: 123 (1.2%). Health: 110 (1.1%). Blockchain and crypto: 101 (1.0%). HR and recruiting: 49 (0.5%). Real estate: 19 (0.2%)Software development2,467 (24.7%)AI and data1,132 (11.3%)Web and mobile apps955 (9.6%)Science and research828 (8.3%)No particular field688 (6.9%)Marketing493 (4.9%)Cybersecurity491 (4.9%)Cloud and IT440 (4.4%)Design and creative work377 (3.8%)Finance357 (3.6%)Personal productivity249 (2.5%)Legal224 (2.2%)Media and publishing209 (2.1%)Games197 (2.0%)E-commerce183 (1.8%)Hardware and engineering179 (1.8%)Sales and customer service129 (1.3%)Education123 (1.2%)Health110 (1.1%)Blockchain and crypto101 (1.0%)HR and recruiting49 (0.5%)Real estate19 (0.2%)
Base: 10,000 skills from 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026. One field for each skill. "No particular field" covers skills of use to anyone.
Science and research leads the fields outside technologyScience and research: 828 (8.3%). Marketing: 493 (4.9%). Design and creative work: 377 (3.8%). Finance: 357 (3.6%). Personal productivity: 249 (2.5%). Legal: 224 (2.2%). Media and publishing: 209 (2.1%). E-commerce: 183 (1.8%). Sales and customer service: 129 (1.3%). Education: 123 (1.2%). Health: 110 (1.1%). HR and recruiting: 49 (0.5%)Science and research828 (8.3%)Marketing493 (4.9%)Design and creative work377 (3.8%)Finance357 (3.6%)Personal productivity249 (2.5%)Legal224 (2.2%)Media and publishing209 (2.1%)E-commerce183 (1.8%)Sales and customer service129 (1.3%)Education123 (1.2%)Health110 (1.1%)HR and recruiting49 (0.5%)Science and research leads the fields outside technologyScience and research: 828 (8.3%). Marketing: 493 (4.9%). Design and creative work: 377 (3.8%). Finance: 357 (3.6%). Personal productivity: 249 (2.5%). Legal: 224 (2.2%). Media and publishing: 209 (2.1%). E-commerce: 183 (1.8%). Sales and customer service: 129 (1.3%). Education: 123 (1.2%). Health: 110 (1.1%). HR and recruiting: 49 (0.5%)Science and research828 (8.3%)Marketing493 (4.9%)Design and creative work377 (3.8%)Finance357 (3.6%)Personal productivity249 (2.5%)Legal224 (2.2%)Media and publishing209 (2.1%)E-commerce183 (1.8%)Sales and customer service129 (1.3%)Education123 (1.2%)Health110 (1.1%)HR and recruiting49 (0.5%)
Base: 10,000 skills from 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026. Fields outside technology only: 3,350 (33.5%) skills.
The most common jobs within fields outside technology (base: 10,000 skills)
FieldMain jobSkills
Science and researchWork with data337
MarketingMarket and sell313
Science and researchResearch and gather information226
Design and creative workMake images, video and audio220
FinanceRun business tasks141
LegalRun business tasks115
FinanceWork with data96
Media and publishingWrite and edit text95
Design and creative workDesign interfaces95
Science and researchWrite and edit text75

Popularity by job and field

Popularity differs little by job: median stars range from 280 to 397Make documents and slides: 397. Deploy and run systems: 356. Write and change code: 349. Market and sell: 348. Review code: 345. Research and gather information: 343. Write and edit text: 342. Test and fix bugs: 334. Run business tasks: 328. Make images, video and audio: 328Make documents and slides397Deploy and run systems356Write and change code349Market and sell348Review code345Research and gather information343Write and edit text342Test and fix bugs334Run business tasks328Make images, video and audio328Popularity differs little by job: median stars range from 280 to 397Make documents and slides: 397. Deploy and run systems: 356. Write and change code: 349. Market and sell: 348. Review code: 345. Research and gather information: 343. Write and edit text: 342. Test and fix bugs: 334. Run business tasks: 328. Make images, video and audio: 328Make documents and slides397Deploy and run systems356Write and change code349Market and sell348Review code345Research and gather information343Write and edit text342Test and fix bugs334Run business tasks328Make images, video and audio328
Median stars of the repositories that hold at least one skill with each main job. Jobs found in fewer than 20 repositories are left out. Base: 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026.
Health and sales skills sit in the most-starred repositoriesHealth: 437. Sales and customer service: 435. Marketing: 368. Web and mobile apps: 365. Science and research: 349. Cybersecurity: 338. Cloud and IT: 333. Blockchain and crypto: 333. No particular field: 329. AI and data: 320Health437Sales and customer service435Marketing368Web and mobile apps365Science and research349Cybersecurity338Cloud and IT333Blockchain and crypto333No particular field329AI and data320Health and sales skills sit in the most-starred repositoriesHealth: 437. Sales and customer service: 435. Marketing: 368. Web and mobile apps: 365. Science and research: 349. Cybersecurity: 338. Cloud and IT: 333. Blockchain and crypto: 333. No particular field: 329. AI and data: 320Health437Sales and customer service435Marketing368Web and mobile apps365Science and research349Cybersecurity338Cloud and IT333Blockchain and crypto333No particular field329AI and data320
Median stars of the repositories that hold at least one skill for each field. Fields found in fewer than 20 repositories are left out. Base: 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026.

AI products named

Claude is named in 22% of popular skills, Codex in 7%Claude: 2,183 (21.8%). Codex: 731 (7.3%). OpenAI: 319 (3.2%). Gemini: 317 (3.2%). Cursor: 308 (3.1%). Copilot: 192 (1.9%). OpenCode: 191 (1.9%). No AI product named: 7,322 (73.2%)Claude2,183 (21.8%)Codex731 (7.3%)OpenAI319 (3.2%)Gemini317 (3.2%)Cursor308 (3.1%)Copilot192 (1.9%)OpenCode191 (1.9%)No AI product named7,322 (73.2%)Claude is named in 22% of popular skills, Codex in 7%Claude: 2,183 (21.8%). Codex: 731 (7.3%). OpenAI: 319 (3.2%). Gemini: 317 (3.2%). Cursor: 308 (3.1%). Copilot: 192 (1.9%). OpenCode: 191 (1.9%). No AI product named: 7,322 (73.2%)Claude2,183 (21.8%)Codex731 (7.3%)OpenAI319 (3.2%)Gemini317 (3.2%)Cursor308 (3.1%)Copilot192 (1.9%)OpenCode191 (1.9%)No AI product named7,322 (73.2%)
Base: 10,000 skills from 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026. A skill can name several products.

1,392 (13.9%) skills name Claude and no other AI product. The Model Context Protocol (MCP) is named in 1,544 (15.4%).

How the skills are built

97% of popular skills give the name and description the format requiresName and description in the header: 9,658 (96.6%). Meets every header rule: 8,749 (87.5%). Longer than 500 lines: 595 (6.0%). Exact copy of another skill: 474 (4.7%). Empty template or stub: 157 (1.6%)Name and description in the header9,658 (96.6%)Meets every header rule8,749 (87.5%)Longer than 500 lines595 (6.0%)Exact copy of another skill474 (4.7%)Empty template or stub157 (1.6%)97% of popular skills give the name and description the format requiresName and description in the header: 9,658 (96.6%). Meets every header rule: 8,749 (87.5%). Longer than 500 lines: 595 (6.0%). Exact copy of another skill: 474 (4.7%). Empty template or stub: 157 (1.6%)Name and description in the header9,658 (96.6%)Meets every header rule8,749 (87.5%)Longer than 500 lines595 (6.0%)Exact copy of another skill474 (4.7%)Empty template or stub157 (1.6%)
Base: 10,000 skills from 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026.
16% of popular skills ship a script; 54% are a single fileSKILL.md is the only file: 4,468 (53.9%). At least one script: 1,322 (16.0%). A references folder: 2,162 (26.1%). An assets or templates folder: 665 (8.0%)SKILL.md is the only file4,468 (53.9%)At least one script1,322 (16.0%)A references folder2,162 (26.1%)An assets or templates folder665 (8.0%)16% of popular skills ship a script; 54% are a single fileSKILL.md is the only file: 4,468 (53.9%). At least one script: 1,322 (16.0%). A references folder: 2,162 (26.1%). An assets or templates folder: 665 (8.0%)SKILL.md is the only file4,468 (53.9%)At least one script1,322 (16.0%)A references folder2,162 (26.1%)An assets or templates folder665 (8.0%)
Base: 8,284 skills kept in a folder of their own, collected 20 November 2026. A skill can appear in more than one row.

The median skill runs to 133 lines, and the median description to 229 characters. 169 (1.7%) skills have exactly the same text as a skill in another repository.

Licences

12% of popular skill repositories carry no licenceMIT: 1,474 (51.9%). Apache-2.0: 513 (18.1%). No licence detected: 353 (12.4%). Licence file not recognised: 313 (11.0%). AGPL-3.0: 81 (2.9%). GPL-3.0: 40 (1.4%)MIT1,474 (51.9%)Apache-2.0513 (18.1%)No licence detected353 (12.4%)Licence file not recognised313 (11.0%)AGPL-3.081 (2.9%)GPL-3.040 (1.4%)12% of popular skill repositories carry no licenceMIT: 1,474 (51.9%). Apache-2.0: 513 (18.1%). No licence detected: 353 (12.4%). Licence file not recognised: 313 (11.0%). AGPL-3.0: 81 (2.9%). GPL-3.0: 40 (1.4%)MIT1,474 (51.9%)Apache-2.0513 (18.1%)No licence detected353 (12.4%)Licence file not recognised313 (11.0%)AGPL-3.081 (2.9%)GPL-3.040 (1.4%)
Base: 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026. Licence as detected by GitHub for the whole repository.

2,025 (71.3%) repositories carry a permissive licence such as MIT or Apache 2.0. A skill in a repository with no licence comes with no stated permission to copy or change it.

Stars and activity

The median popular skill repository has 309 stars100 to 199: 990 (34.8%). 200 to 499: 795 (28.0%). 500 to 999: 429 (15.1%). 1,000 to 2,999: 383 (13.5%). 3,000 to 10,000: 244 (8.6%)100 to 199990 (34.8%)200 to 499795 (28.0%)500 to 999429 (15.1%)1,000 to 2,999383 (13.5%)3,000 to 10,000244 (8.6%)The median popular skill repository has 309 stars100 to 199: 990 (34.8%). 200 to 499: 795 (28.0%). 500 to 999: 429 (15.1%). 1,000 to 2,999: 383 (13.5%). 3,000 to 10,000: 244 (8.6%)100 to 199990 (34.8%)200 to 499795 (28.0%)500 to 999429 (15.1%)1,000 to 2,999383 (13.5%)3,000 to 10,000244 (8.6%)
Base: 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026.
62% of popular skill repositories changed in the last 30 daysWithin 30 days: 1,772 (62.4%). 31 to 90 days: 459 (16.2%). 91 days to a year: 603 (21.2%). More than a year: 7 (0.2%)Within 30 days1,772 (62.4%)31 to 90 days459 (16.2%)91 days to a year603 (21.2%)More than a year7 (0.2%)62% of popular skill repositories changed in the last 30 daysWithin 30 days: 1,772 (62.4%). 31 to 90 days: 459 (16.2%). 91 days to a year: 603 (21.2%). More than a year: 7 (0.2%)Within 30 days1,772 (62.4%)31 to 90 days459 (16.2%)91 days to a year603 (21.2%)More than a year7 (0.2%)
Time from the last commit on the main branch to 20 November 2026. Base: 2,841 public GitHub repositories with 100 to 10,000 stars, collected 20 November 2026.

1,992 (70.1%) of the repositories were created in 2026, and 25 (0.9%) are archived.

Repositories over 10,000 stars

424 repositories with more than 10,000 stars hold 26,648 SKILL.md files. They are kept out of the 10,000 so that a few very large collections do not decide the figures. A separate sample of 500 skills from 127 of them was sorted the same way: build and run ai agents is the main job of 74 (14.8%).

Main job in repositories over 10,000 stars (base: 500 skills)
Main jobSkillsShare
Build and run AI agents7414.8%
Check security5511.0%
Write and change code5010.0%
Market and sell326.4%
Deploy and run systems326.4%
Work with data326.4%

Method

  • Repositories. GitHub repository search by topic and keyword (agent skills, Claude skills, Claude Code, Codex, Gemini CLI and others), and GitHub code search for files named SKILL.md, gave 44,245 candidate repositories. Forks were left out. 10,182 repositories with 100 to 10,000 stars held at least one SKILL.md file, 359,868 in all.
  • Skills. Every file named SKILL.md in each repository's full file list. To keep a few large collections from deciding the figures, each repository gave at most 40 skills, chosen in a fixed random order. 10,000 skills were then taken in the same order. Unreadable files and files that only point to another SKILL.md were replaced by the next in order.
  • Job and field. Each skill was sorted into one of 17 jobs and one of 22 fields by a language model (gemini-3-8-flash) from its name, description, opening text and repository description, using fixed lists. The same model marked empty templates and stubs.
  • Header. Read from the block between two lines of three hyphens at the top of the file. "Meets every header rule" means a name in lower-case letters, digits and hyphens of up to 64 characters, equal to the folder name, and a description of 1 to 1,024 characters, as the Agent Skills specification sets out.
  • Scripts. A skill ships a script when its folder, or a folder directly inside it, holds a file ending in .py, .sh, .bash, .zsh, .ps1, .bat, .cmd, .js, .mjs, .cjs, .ts, .rb, .pl, .php, .lua, .r. Skills at the top of a repository are left out of this count.
  • AI products. A search of the full text for each product name, without regard to capitals.
  • Licence, stars and activity. As GitHub reports them for the repository on the day of collection.

Check of a sample

400 skills, every 25th in a fixed random order, were read in full by a second reader (a second AI model, Claude Sonnet, not a person) and compared with the recorded values.

Disagreements found (base: 400 skills)
ValueWrongError rate
Main job00.0%
Field20.5%
Template or stub20.5%
Name and description00.0%
Ships scripts00.0%
AI products named51.3%

In all, 9 (0.4%) of 2,400 values checked were wrong. Figures are given as counted, without correction.

Limits

  • GitHub search finds repositories by their topics, names, descriptions and recent activity. Repositories that hold skills but are described in other terms may be missing.
  • Stars belong to a repository, not to a skill. Installs and downloads are not public, so use of a single skill was not measured.
  • The job and field come from a language model; the check above gives their error rate.
  • No skill was run, and no script was read or tested for safety.
  • The figures describe one day. Repositories change, and the next collection will differ.

Data and citation

One record for each of the 10,000 skills, with no repository, owner, skill name or file path: repositories are numbered, stars are given in bands and the last commit by month.

Published by Knacco, a library of ready-to-use AI skills, prompt packs and workflows. The data, figures and charts are free to share and adapt under the CC BY 4.0 licence, with credit and a link to this page. Corrections are received through the contact page.

Agent skills study 2026: what 10,000 popular skills do. Data collected 20 November 2026. https://knacco.com/en/research/agent-skills-study