Agent skill
telegram
Quellcode ansehen: ninehills/skills
Installation
npx skills add ninehills/skills --skill telegram23
Installationen
EU-hosted inference API
Power your AI agent skills with open-source models.
Drop-in OpenAI-compatible API. No data leaves Europe.
MiniMax
MiniMax M3.1 Flash
$0.30 / $1.20
per M tokens
Z.ai
GLM 5.3 Flash
$0.20 / $0.60
per M tokens
MoonshotAI
Kimi K3.1 Pro
$3.00 / $14.00
per M tokens
DeepSeek
DeepSeek V4.1 Flash
$0.40 / $1.42
per M tokens
Skills Collection
A curated collection of agent skills for Claude Code and compatible agent platforms.
CLI 工具:skills-manager
使用 skills-manager 脚本将场景中的技能通过软连接安装到各 Agent 平台的 skills 目录。
使用方式
# 列出所有可用场景
python3 skills-manager scenarios list
# 安装指定场景(将技能软连接到默认目标目录)
python3 skills-manager scenarios install Common默认安装目标
软连接会安装到以下目录(自动创建,已存在则跳过):
| 平台 | 目标目录 |
|---|---|
| Codex | ~/.codex/skills |
| Pi | ~/.pi/agent/skills |
| Claude Code | ~/.claude/skills |
| Hermes | ~/.hermes/skills |
软连接清理
每次执行 install 时,脚本会自动清理目标目录中指向当前仓库、但不在当前场景配置中的无效软连接,确保安装目录保持整洁。
Quick Reference
| Skill | Purpose | Use When |
|---|---|---|
| agent-browser | Browser automation | Interacting with websites, scraping, testing web apps |
| brainstorming | Design exploration | Starting any creative work, building features, components |
| find-skills | Skill discovery | Looking for tools, asking "how do I do X" |
| frontend-design | Visual interface creation | Building web components, pages, landing pages |
| frontend-skill | Premium UI development | Art direction, award-level composition |
| pua | Performance coaching | User frustration, repeated failures, debugging |
| skill-creator | Skill development | Creating or improving skills |
| ui-ux-pro-max | Professional design | High-end UI/UX work |
| vercel-react-best-practices | React optimization | Writing, reviewing React/Next.js code |
| web-design-guidelines | UI code review | Reviewing UI for accessibility and best practices |
| writing-plans | Implementation planning | Multi-step tasks with clear specs |
Skill Details
agent-browser
Purpose: Browser automation CLI for AI agents.
Use Cases:
- Navigate websites and interact with pages
- Fill forms, click buttons, take screenshots
- Scrape data, test web applications
- Handle authentication flows
Key Commands:
agent-browser open <url>— Navigate to URLagent-browser snapshot -i— Get interactive elements with refsagent-browser click @e1/agent-browser fill @e2 "text"— Interact via refsagent-browser screenshot— Capture page images
brainstorming
Purpose: Collaborative design exploration before implementation.
When to Use: Before any creative work — creating features, building components, adding functionality, or modifying behavior.
Process:
- Explore project context
- Ask clarifying questions (one at a time)
- Propose 2-3 approaches with trade-offs
- Present design sections for approval
- Write design doc to
docs/superpowers/specs/ - Invoke
writing-plansfor implementation
Hard Gate: Do NOT write code until design is approved.
find-skills
Purpose: Discover and install skills from the open agent skills ecosystem.
When to Use:
- "How do I do X?"
- "Find a skill for X"
- "Is there a skill that can..."
Key Commands:
npx skills find [query]— Search for skillsnpx skills add <package>— Install a skill- Browse: https://skills.sh/
frontend-design
Purpose: Create distinctive, production-grade frontend interfaces.
When to Use: Building web components, pages, artifacts, posters, or applications.
Design Principles:
- Bold aesthetics — Choose an extreme direction (minimalist, maximalist, retro-futuristic, etc.)
- Typography — Distinctive, characterful font choices
- Motion — Well-orchestrated animations, scroll-triggered effects
- Spatial composition — Unexpected layouts, asymmetry, overlap
Avoid: Generic AI aesthetics (Inter font, purple gradients, predictable layouts)
frontend-skill
Purpose: Ship premium, award-level interfaces with art direction and restraint.
When to Use: When quality depends on art direction, hierarchy, imagery, and motion.
Defaults:
- Start with composition, not components
- Full-bleed hero, no card grids
- Brand-first hierarchy
- Two typefaces max, one accent color
- Cardless layouts preferred
Litmus Checks:
- Is the brand unmistakable in the first screen?
- Can the page be understood by scanning headlines only?
- Would the design feel premium without decorative shadows?
pua
Purpose: High-agency problem-solving with performance coaching.
When to Use:
- User frustration or repeated failures (2+)
- Passive behavior or quality complaints
- Triggers: "try harder", "figure it out", "stop giving up", "you broke it"
Methodology Routing:
| Task Type | Flavor | Core Method |
|---|---|---|
| Debug/Bug | Huawei | RCA root cause + Blue Team self-attack |
| New Feature | Musk | The Algorithm: question → delete → simplify → accelerate → automate |
| Code Review | Jobs | Subtraction priority + pixel-perfect + DRI |
| Research | Baidu | Search is primary productivity |
| Architecture | Amazon | Working Backwards + 6-Pager |
| Performance | ByteDance | A/B Test + data-driven |
skill-creator
Purpose: Create, modify, and measure skill performance.
When to Use:
- Creating a skill from scratch
- Editing or optimizing existing skills
- Running evals to test skills
- Benchmarking skill performance
Workflow:
- Capture intent (what, when, output format)
- Interview and research edge cases
- Write
SKILL.mdwith YAML frontmatter - Create test cases in
evals/evals.json - Run with-skill vs baseline subagents
- Review, iterate, package
ui-ux-pro-max
Purpose: Professional UI/UX design for high-end work.
When to Use: Complex UI/UX projects requiring professional design expertise.
vercel-react-best-practices
Purpose: React and Next.js performance optimization from Vercel Engineering.
When to Use:
- Writing new React components or Next.js pages
- Implementing data fetching
- Reviewing code for performance
- Optimizing bundle size or load times
Rule Categories (65 rules):
| Priority | Category | Prefix |
|---|---|---|
| 1 | Eliminating Waterfalls | async- |
| 2 | Bundle Size Optimization | bundle- |
| 3 | Server-Side Performance | server- |
| 4 | Client-Side Data Fetching | client- |
| 5 | Re-render Optimization | rerender- |
| 6 | Rendering Performance | rendering- |
| 7 | JavaScript Performance | js- |
| 8 | Advanced Patterns | advanced- |
web-design-guidelines
Purpose: Review UI code for Web Interface Guidelines compliance.
When to Use:
- "Review my UI"
- "Check accessibility"
- "Audit design"
- "Review UX"
Process:
- Fetch latest guidelines from source
- Read specified files
- Check against all rules
- Output findings in
file:lineformat
writing-plans
Purpose: Write comprehensive implementation plans for multi-step tasks.
When to Use: You have a spec or requirements for a multi-step task, before touching code.
Output: Plans saved to docs/superpowers/plans/YYYY-MM-DD-<feature-name>.md
Structure:
- Bite-sized tasks (2-5 minutes each)
- Exact file paths
- Complete code in every step
- Exact commands with expected output
- DRY, YAGNI, TDD, frequent commits
Installation
Install individual skills using the Skills CLI:
npx skills add owner/repo@skill-nameOr install from this repository:
# Clone the repository
git clone https://github.com/ninehills/skills.git
cd skills
# Install a specific skill
npx skills add ./agent-browser -gSkill Structure
Each skill follows this structure:
skill-name/
├── SKILL.md # Required: skill definition with YAML frontmatter
├── references/ # Optional: documentation loaded as needed
├── scripts/ # Optional: executable code for deterministic tasks
└── assets/ # Optional: templates, icons, fontsLicense
See individual skill directories for license information.
Verwandte Skills
Mehr aus dieser Quelle: ninehills/skills
So verwenden Sie diesen Skill
- 1
Install telegram by running
npx skills add ninehills/skills --skill telegramin your project directory. Führen Sie den obigen Installationsbefehl in Ihrem Projektverzeichnis aus. Die Skill-Datei wird von GitHub heruntergeladen und in Ihrem Projekt platziert. - 2
Keine Konfiguration erforderlich. Ihr KI-Agent (Claude Code, Cursor, Windsurf usw.) erkennt installierte Skills automatisch und nutzt sie als Kontext bei der Code-Generierung.
- 3
Der Skill verbessert das Verständnis Ihres Agenten für telegram, und hilft ihm, etablierte Muster zu befolgen, häufige Fehler zu vermeiden und produktionsreifen Code zu erzeugen.
Was Sie erhalten
Skills sind Klartext-Anweisungsdateien — kein ausführbarer Code. Sie kodieren Expertenwissen über Frameworks, Sprachen oder Tools, das Ihr KI-Agent liest, um seine Ausgabe zu verbessern. Das bedeutet null Laufzeit-Overhead, keine Abhängigkeitskonflikte und volle Transparenz: Sie können jede Anweisung vor der Installation lesen und prüfen.
Kompatibilität
Dieser Skill funktioniert mit jedem KI-Coding-Agenten, der das skills.sh-Format unterstützt, einschließlich Claude Code (Anthropic), Cursor, Windsurf, Cline, Aider und anderen Tools, die projektbezogene Kontextdateien lesen. Skills sind auf Transportebene framework-agnostisch — der Inhalt bestimmt, für welche Sprache oder welches Framework er gilt.
Data sourced from the skills.sh registry and GitHub. Install counts and security audits are updated regularly.