--- title: Top Skills for Claude API slug: claude-api type: agent vendor: Anthropic native: true canonical: https://top-agent-skills.com/skills-for-claude-api skills: 13 rubric: 1.0 updated: 2026-07-15 --- # Top Skills for Claude API For the Claude API the practical picks are pdf, web-artifacts-builder and frontend-design — Skills whose whole job is file in, file out, which is what survives a programmatic call. This is the thinnest surface in the catalogue by verified count, and that is the honest signal: most Skills are written for an interactive agent with a shell, not for a request/response API. Use pre-built skill_id values or upload your own via /v1/skills. 13 Skills verified on Claude API. Vendor: Anthropic. Runs SKILL.md natively. Compatibility here lists only Skills actually tested on this agent or shipping a publisher-documented path for it — never assumed. Vendor install docs: https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview ## The ranking ### 1. frontend-design — score 92/100 (official-anthropic) Escape generic AI-generated UIs. Forces a bold, distinctive design direction before any code is written. - Best for: Greenfield landing pages, marketing sites, and any UI that should feel intentional rather than templated. - Not ideal for: Ports of an existing design system — frontend-design pushes a bold direction that may overwrite established tokens. - Publisher: Anthropic · MIT - Verified on: claude-code, claude-ai, claude-api, cursor, codex, gemini-cli, antigravity ```bash npx skills add github.com/anthropics/skills --skill frontend-design ``` Full entry: https://top-agent-skills.com/skill/frontend-design (https://top-agent-skills.com/skill/frontend-design.md) ### 2. composio — score 55/100 (verified-org) 1,000+ SaaS integrations as one skill. OAuth lifecycle, scoped credentials, structured actions. - Best for: Any agent that needs to act on real SaaS systems — Slack, Notion, GitHub, Stripe, Salesforce. - Not ideal for: Read-only research where MCPs would be lighter. - Publisher: Composio · Apache-2.0 - Verified on: claude-code, claude-ai, claude-api, cursor, codex, gemini-cli, antigravity ```bash npx skills add composiohq/skills ``` Full entry: https://top-agent-skills.com/skill/composio (https://top-agent-skills.com/skill/composio.md) ### 3. web-artifacts-builder — score 52/100 (official-anthropic) Self-contained HTML artifacts with React + shadcn/ui — no build step, just paste and run. - Best for: Demos, prototypes, and one-off interactive artifacts. - Publisher: Anthropic · MIT - Verified on: claude-code, claude-ai, claude-api, cursor, codex ```bash npx skills add github.com/anthropics/skills --skill web-artifacts-builder ``` Full entry: https://top-agent-skills.com/skill/web-artifacts-builder (https://top-agent-skills.com/skill/web-artifacts-builder.md) ### 4. pdf — score 52/100 (official-anthropic) Extract, fill, merge. PDF manipulation with bundled scripts so Claude does not waste tokens generating them. - Best for: Bulk PDF text/table extraction, form filling, merging. - Publisher: Anthropic · MIT - Verified on: claude-code, claude-ai, claude-api, cursor, codex ```bash npx skills add github.com/anthropics/skills --skill pdf ``` Full entry: https://top-agent-skills.com/skill/pdf (https://top-agent-skills.com/skill/pdf.md) ### 5. accessibility-review — score 52/100 (official-anthropic) Pre-handoff a11y audit that catches what design QA misses. - Best for: Reviewing a design or live page against WCAG 2.1 AA before engineering handoff or production deploy. - Not ideal for: Replacing manual screen-reader testing with assistive tech in real environments. - Publisher: Anthropic · MIT - Verified on: claude-code, claude-ai, claude-api, cursor, codex, antigravity ```bash claude plugin install design ``` Full entry: https://top-agent-skills.com/skill/accessibility-review (https://top-agent-skills.com/skill/accessibility-review.md) ### 6. stripe-payments — score 51/100 (verified-org) Stripe done right. Checkout, subscriptions, webhooks, idempotency, metered billing. Full entry: https://top-agent-skills.com/skill/stripe-payments (https://top-agent-skills.com/skill/stripe-payments.md) ### 7. firecrawl — score 49/100 (verified-org) Web scraping, search, and browser automation built to be agent-friendly. Full entry: https://top-agent-skills.com/skill/firecrawl (https://top-agent-skills.com/skill/firecrawl.md) ### 8. notion-api-skill — score 49/100 (verified-org) Notion API: page/database CRUD, block hierarchies, rich text, schema sync. Full entry: https://top-agent-skills.com/skill/notion-api-skill (https://top-agent-skills.com/skill/notion-api-skill.md) ### 9. brand-guidelines — score 48/100 (official-anthropic) Lock generated UIs to a brand. Anthropic's own guidelines, formalized as a skill. Full entry: https://top-agent-skills.com/skill/brand-guidelines (https://top-agent-skills.com/skill/brand-guidelines.md) ### 10. docx — score 48/100 (official-anthropic) Word documents end-to-end. Create, edit, comment, track-change. Full entry: https://top-agent-skills.com/skill/docx (https://top-agent-skills.com/skill/docx.md) ### 11. xlsx — score 48/100 (official-anthropic) Excel spreadsheets with real formulas, charts, and formatting — not flat CSV exports. Full entry: https://top-agent-skills.com/skill/xlsx (https://top-agent-skills.com/skill/xlsx.md) ### 12. pptx — score 48/100 (official-anthropic) PowerPoint decks with layouts, charts, speaker notes — programmatic from prose. Full entry: https://top-agent-skills.com/skill/pptx (https://top-agent-skills.com/skill/pptx.md) ### 13. valyu — score 46/100 (verified-org) Web search + 36 authoritative data sources. SEC, PubMed, ChEMBL, FRED — agent-grade research. Full entry: https://top-agent-skills.com/skill/valyu (https://top-agent-skills.com/skill/valyu.md) ## Common questions **How do I install a Skill via the Claude API?** POST your Skill zip to /v1/skills. You receive a skill_id. Reference it in your messages request — the model loads the Skill on demand. **Which pre-built Skills are available?** Anthropic publishes skill_id values for docx, pdf, pptx, xlsx, and several others. The full list is in the platform.claude.com documentation. **Do Skills work with prompt caching?** Yes — Skill definitions are cached as part of the prompt prefix, so repeated invocations within the 5-minute TTL hit the cache cleanly. **Are Skills compatible with the Batch API?** Yes — every request in a batch can reference Skills the same way single requests do. This is the right target for high-volume document processing. **Which Claude model should I use?** Claude Opus 4.7 (claude-opus-4-7) for complex Skills, Claude Sonnet 4.6 (claude-sonnet-4-6) for cost-efficient routine work, Claude Haiku 4.5 for lightweight Skills. Skill performance depends heavily on the model. --- _Top Agent Skills. Machine-readable twin of https://top-agent-skills.com/skills-for-claude-api._ _JSON: https://top-agent-skills.com/skills-for-claude-api.json · whole catalog: https://top-agent-skills.com/skills.json (CC BY 4.0)_