# State of Agent Skills 2026

Published: 2026-07-10
Generated: 2026-07-10T14:09:59.307Z
Canonical URL: https://www.openagentskill.com/reports/state-of-agent-skills-2026

## Scope
- Indexed candidates: 20,466
- Analyzed candidates: 985
- Strict agent-skill candidates: 71
- Agent workflow candidates: 120
- Domain workflow and ecosystem tools: 294
- Lists and collections: 19
- Generic candidates: 481
- Maintained within 180 days: 984
- Known license: 798
- High trust: 985
- Agent Proven skills: 1
- Reported agent outcomes: 1

## Findings
- 7.2% of the analyzed sample has strict Agent Skill evidence in its repository name, README, description, or topics.
- 12.2% describes an agent workflow or agent product without strict reusable-skill evidence.
- 29.8% is better described as a domain workflow tool or ecosystem dependency than a strict agent skill.
- 1.9% is a list or collection rather than one installable skill.
- 99.9% was pushed within 180 days of report generation.
- 81% has a declared license that is not marked unknown.
- 100% reaches the current 72-point high-trust threshold in this quality-ranked sample; Trust remains a metadata signal, not proof of task success.
- 1 analyzed skill has at least one reported agent outcome, covering 1 outcome in total.

## Leading skill-specific candidates
1. Crawl4AI — 72K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/crawl4ai
2. Taste Skill — 61K stars, 95/100 trust, safe_to_try, https://www.openagentskill.com/skills/leonxlnx-taste-skill
3. Code Review — 164K stars, 94/100 trust, safe_to_try, https://www.openagentskill.com/skills/mattpocock-code-review
4. Open Design — 77K stars, 94/100 trust, safe_to_try, https://www.openagentskill.com/skills/nexu-io-open-design
5. Marketingskills — 37K stars, 94/100 trust, safe_to_try, https://www.openagentskill.com/skills/coreyhaines31-marketingskills
6. Nanoclaw — 30K stars, 94/100 trust, safe_to_try, https://www.openagentskill.com/skills/nanocoai-nanoclaw
7. Last30days Skill — 51K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/mvanhorn-last30days-skill
8. Scientific Agent Skills — 31K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/k-dense-ai-scientific-agent-skills
9. Eigent — 15K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/eigent-ai-eigent
10. Ponytail — 79K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/dietrichgebert-ponytail
11. Cherry Studio — 48K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/cherryhq-cherry-studio
12. OpenMontage — 36K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/calesthio-openmontage
13. Apple Design — 7.9K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/emilkowalski-apple-design
14. SkillSpector — 13K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/nvidia-skillspector
15. Pm Skills — 22K stars, 93/100 trust, safe_to_try, https://www.openagentskill.com/skills/phuryn-pm-skills
16. Huashu Design — 19K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/alchaincyf-huashu-design
17. Guizang Ppt Skill — 17K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/op7418-guizang-ppt-skill
18. Claude Seo — 8.9K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/agricidaniel-claude-seo
19. Review Animations — 7.9K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/emilkowalski-review-animations
20. Animation Vocabulary — 7.9K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/emilkowalski-animation-vocabulary
21. Visual Explainer — 9.2K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/nicobailon-visual-explainer
22. Emil Design Engineering — 7.9K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/emilkowalski-emil-design-eng
23. Cognee — 28K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/topoteretes-cognee
24. Frontend Slides — 22K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/zarazhangrui-frontend-slides
25. Ui Ux Pro Max Skill — 104K stars, 92/100 trust, safe_to_try, https://www.openagentskill.com/skills/nextlevelbuilder-ui-ux-pro-max-skill

## Methodology
All public records with ai_review_approved=true in the OpenAgentSkill registry. MCP-only records are excluded by the registry read layer.
The highest-ranked 985 approved candidates by the current quality ordering. The sample is a product audit, not a random statistical sample.
- Limitation: GitHub stars measure repository adoption, not agent task success
- Limitation: Candidate listings are not equivalent to verified maintainer claims
- Limitation: A public metadata audit is not a substitute for source review or sandbox execution
- Limitation: Agent Proven evidence remains sparse until agents report real outcomes
- Limitation: Repository metadata can change after this report is generated