Open research · 2026 edition
State of Agent Skills 2026.
A transparent look at what is actually inside an agent-skill registry: which records look like reusable skills, which are maintained, which declare a license, what can be installed, and how little real outcome evidence the ecosystem still has.
28,456
Indexed candidates
62
Analyzed sample
79%
Strict agent skills
0
Agent Proven
Executive finding
Popularity is not readiness.
GitHub adoption is useful, but it does not prove task fit, safe installation, or a successful agent run. This report keeps those signals separate.
- 0179% of the analyzed sample has strict Agent Skill evidence in its repository name, README, description, or topics.
- 026.5% describes an agent workflow or agent product without strict reusable-skill evidence.
- 0312.9% is better described as a domain workflow tool or ecosystem dependency than a strict agent skill.
- 041.6% is a list or collection rather than one installable skill.
- 0590.3% was pushed within 180 days of report generation.
- 0685.5% has a declared license that is not marked unknown.
- 0798.4% reaches the current 72-point high-trust threshold in this quality-ranked sample; Trust remains a metadata signal, not proof of task success.
- 080 analyzed skills have at least one reported agent outcome, covering 0 outcomes in total.
Readiness signals
What the analyzed sample can support.
- Maintained within 180 days
- 5690%
- Known license
- 5385%
- Install path available
- 00%
- 500+ GitHub stars
- 5284%
- 72+ Trust Score
- 6198%
- Safe-to-try audit
- 2540%
Domain coverage
Where the candidate supply is concentrated.
Track assignment uses category, task, use-case, and repository text signals. It is a discovery aid, not a claim that every record is production ready.
| Track | Candidates | Share | Maintained | Known license | High trust |
|---|---|---|---|---|---|
| Finance and quant workflows | 25 | 40.3% | 20 | 22 | 25 |
| Design and creative production | 16 | 25.8% | 15 | 16 | 15 |
| Coding and developer agents | 11 | 17.7% | 11 | 9 | 11 |
| Research and knowledge work | 5 | 8.1% | 5 | 4 | 5 |
| Football and World Cup analytics | 2 | 3.2% | 2 | 0 | 2 |
| Marketing and growth automation | 2 | 3.2% | 2 | 2 | 2 |
| General agent automation | 1 | 1.6% | 1 | 0 | 1 |
Evidence shortlist
Leading strict skill candidates.
Ordered by reported outcomes, Trust Score, skill-likeness, and GitHub adoption. This is a shortlist for evaluation, not an automatic-install endorsement.
| Skill | Track | Stars | Trust | Audit | Risk | Outcomes |
|---|---|---|---|---|---|---|
| Taste Skill: Anti-Slop Frontend Leonxlnx/taste-skill | Design and creative production | 66K | 93 | 94 | Safe to try | 0 |
| Vercel React Best Practices vercel-labs/agent-skills | Coding and developer agents | 29K | 93 | 94 | Safe to try | 0 |
| Frontend Design anthropics/skills | Design and creative production | 163K | 90 | 92 | Safe to try | 0 |
| Figma Implement Design openai/skills | Design and creative production | 24K | 90 | 92 | Safe to try | 0 |
| Anthropic Brand Guidelines anthropics/skills | Design and creative production | 163K | 90 | 92 | Safe to try | 0 |
| Web Design Guidelines vercel-labs/agent-skills | Design and creative production | 29K | 90 | 92 | Safe to try | 0 |
| Canvas Design anthropics/skills | Design and creative production | 163K | 89 | 91 | Safe to try | 0 |
| Webapp Testing anthropics/skills | Coding and developer agents | 163K | 88 | 91 | Safe to try | 0 |
| Deploy to Vercel vercel-labs/agent-skills | Coding and developer agents | 29K | 87 | 90 | Safe to try | 0 |
| Playwright Browser Skill openai/skills | Coding and developer agents | 24K | 85 | 89 | Safe to try | 0 |
| Equity Research Earnings Analysis anthropics/financial-services | Finance and quant workflows | 34K | 83 | 90 | Needs review | 0 |
| Portfolio Monitoring anthropics/financial-services | Finance and quant workflows | 34K | 83 | 89 | Needs review | 0 |
| Seedance 2.0 Prompt Writing dexhunter/seedance2-skill | Design and creative production | 2.9K | 83 | 86 | Safe to try | 0 |
| 3-Statement Financial Model anthropics/financial-services | Finance and quant workflows | 34K | 83 | 90 | Needs review | 0 |
| Earnings Preview himself65/finance-skills | Finance and quant workflows | 3.1K | 81 | 89 | Needs review | 0 |
| US Stock Analysis tradermonty/claude-trading-skills | Finance and quant workflows | 2.6K | 81 | 89 | Needs review | 0 |
| Earnings Calendar tradermonty/claude-trading-skills | Finance and quant workflows | 2.6K | 81 | 89 | Needs review | 0 |
| Company Valuation himself65/finance-skills | Finance and quant workflows | 3.1K | 81 | 89 | Needs review | 0 |
| Market Environment Analysis tradermonty/claude-trading-skills | Finance and quant workflows | 2.6K | 81 | 89 | Needs review | 0 |
| Aaron Marketing Skills aaron-he-zhu/aaron-marketing-skills | Marketing and growth automation | 2.3K | 80 | 88 | Safe to try | 0 |
| Vox Director Alisa0808/vox-director | Design and creative production | 326 | 80 | 89 | Safe to try | 0 |
| Market News Analyst tradermonty/claude-trading-skills | Finance and quant workflows | 2.6K | 79 | 88 | Needs review | 0 |
| Tech Earnings Deep Dive star23/Day1Global-Skills | Finance and quant workflows | 1.0K | 79 | 87 | Needs review | 0 |
| David Ondrej Skills davidondrej/skills | Coding and developer agents | 968 | 79 | 88 | Safe to try | 0 |
| Simons Quant xuboyuebobb/investorskills | Finance and quant workflows | 811 | 79 | 88 | Needs review | 0 |
Methodology
Readable before repeatable.
Published Jul 10, 2026. Data refreshed Oct 11, 2026.
Population and sample
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 62 approved candidates by the current quality ordering. The sample is a product audit, not a random statistical sample.
Inclusion and measurement
- •Public GitHub repository or repository URL is available
- •Registry AI review is approved
- •MCP-only server records are excluded
- •Skill-likeness is measured separately from repository popularity
- •Trust and audit scores use public repository metadata and declared install information
Limitations
- •GitHub stars measure repository adoption, not agent task success
- •Candidate listings are not equivalent to verified maintainer claims
- •A public metadata audit is not a substitute for source review or sandbox execution
- •Agent Proven evidence remains sparse until agents report real outcomes
- •Repository metadata can change after this report is generated
Cite this report
OpenAgentSkill. “State of Agent Skills 2026.” Published July 10, 2026. Dataset and methodology available at https://www.openagentskill.com/reports/state-of-agent-skills-2026.