@hasna/skills 0.1.33 → 0.1.35

package/skills/deepresearch/SKILL.md

@@ -1,50 +1,41 @@
 ---
 name: deepresearch
-description: Perform deep research on any topic using Exa.ai for parallel semantic search and Claude/OpenAI for synthesis. Use when you need comprehensive research reports with citations, analysis of topics from multiple angles, or thorough investigation of technical subjects.
+description: Run hosted deep research with parallel search, synthesis, citations, source notes, and downloadable report artifacts.
 ---
 
 # Deep Research
 
-Agentic deep research skill using Exa.ai for search and Claude/OpenAI for synthesis.
+Deep Research is a hosted premium skill. The OSS package exposes public
+metadata and usage guidance only; search providers, model routing, prompts,
+worker orchestration, and credentials stay server-side.
 
-## Invocation
+## Usage
 
-```
-/deepresearch <topic>
+```bash
+skills setup --mode hosted
+skills auth login
+skills run deepresearch "Best practices for building production RAG systems" --depth normal
+skills runs status <run-id>
+skills exports download <run-id>
 ```
 
 ## Options
 
 | Option | Description | Default |
-|--------|-------------|---------|
-| `--depth <level>` | Research depth: quick (6 queries), normal (15), deep (30) | normal |
-| `--model <provider>` | LLM: claude or openai | claude |
-| `--output <path>` | Custom output path | ~/.skills/deepresearch/exports/ |
-| `--json` | Also save sources as JSON | false |
-| `--no-firecrawl` | Skip deep scraping | false |
-
-## Examples
-
-```bash
-# Quick overview of a topic
-/deepresearch "What is RAG?" --depth quick
-
-# Standard research with Claude
-/deepresearch "Best practices for building production ML systems"
-
-# Deep research with OpenAI, custom output
-/deepresearch "Compare Next.js vs Remix" --depth deep --model openai --output ./research.md
-```
+| --- | --- | --- |
+| `--depth <level>` | Research depth: `quick`, `normal`, or `deep` | `normal` |
+| `--output <path>` | Requested output path inside the exported artifact bundle | hosted export |
+| `--json` | Request raw source metadata in the exported bundle | `false` |
 
 ## Requirements
 
-- `EXA_API_KEY` - Required for search
-- `ANTHROPIC_API_KEY` - Required for Claude synthesis
-- `OPENAI_API_KEY` - Required for OpenAI synthesis
-- `FIRECRAWL_API_KEY` - Optional for deep scraping
+- Authenticate with `skills auth login` or provide `SKILLS_API_KEY`.
+- Provider credentials are managed by the hosted runtime and are not part of the
+  OSS package.
 
-## Output
+## Outputs
 
-Reports are saved to `~/.skills/deepresearch/exports/` with format:
-- `report-{topic-slug}-{timestamp}.md` - Full research report
-- `sources-{topic-slug}-{timestamp}.json` - Raw sources (if --json)
+- Research report
+- Source notes
+- Citation metadata
+- Run manifest

package/skills/image/README.md

@@ -1,195 +1,25 @@
-# Image Generation Skill
+# Image Generation
 
-A Claude Code skill for generating images using multiple AI providers: OpenAI DALL-E 3, Google Imagen 3, and xAI Aurora.
-
-## Features
-
-- Multiple AI providers in one unified interface
-- Clean, type-safe TypeScript implementation
-- Simple CLI interface
-- Native Bun runtime (no external dependencies)
-- Support for custom models and sizes
-- Automatic image download and saving
-
-## Installation
-
-```bash
-cd image
-bun install
-```
-
-## Configuration
-
-Set up your API keys as environment variables:
-
-```bash
-# OpenAI
-export OPENAI_API_KEY="your-openai-api-key"
-
-# Google Gemini
-export GEMINI_API_KEY="your-gemini-api-key"
-export GOOGLE_PROJECT_ID="your-google-project-id"
-
-# xAI
-export XAI_API_KEY="your-xai-api-key"
-```
+Hosted image generation skill with provider-cost pricing and downloadable image
+artifacts.
 
 ## Usage
 
-### Basic Commands
-
 ```bash
-# OpenAI DALL-E 3
-bun run src/index.ts generate --provider openai --prompt "a cat" --output ./cat.png
-
-# Google Imagen 3
-bun run src/index.ts generate --provider google --prompt "a dog" --output ./dog.png
-
-# xAI Aurora
-bun run src/index.ts generate --provider xai --prompt "a bird" --output ./bird.png
-```
-
-### Advanced Options
-
-```bash
-# Custom size with OpenAI
-bun run src/index.ts generate \
-  --provider openai \
-  --prompt "a futuristic cityscape" \
-  --output ./city.png \
-  --size 1792x1024
-
-# Custom model
-bun run src/index.ts generate \
-  --provider openai \
-  --prompt "abstract art" \
-  --output ./art.png \
-  --model dall-e-3
-
-# Google Imagen with aspect ratio
-bun run src/index.ts generate \
-  --provider google \
-  --prompt "mountain landscape" \
-  --output ./mountain.png \
-  --size 16:9
+skills setup --mode hosted
+skills auth login
+skills run image "editorial product photo on a white sweep"
 ```
 
-### Short Flags
-
-```bash
-bun run src/index.ts generate \
-  -p openai \
-  --prompt "a sunset" \
-  -o ./sunset.png \
-  -s 1024x1024
-```
-
-## Provider Details
-
-### OpenAI DALL-E 3
-
-- **Endpoint**: `https://api.openai.com/v1/images/generations`
-- **Models**:
-  - `dall-e-3` (default) - Latest DALL-E model
-  - `gpt-image-1` - GPT-4o image generation
-- **Sizes**:
-  - `1024x1024` (default, square)
-  - `1792x1024` (landscape)
-  - `1024x1792` (portrait)
-- **Features**: High-quality, detailed images with prompt revision
-
-### Google Imagen 3
-
-- **Endpoint**: Vertex AI REST API
-- **Model**: `imagen-3.0-generate-001`
-- **Aspect Ratios**:
-  - `1:1` (default, square)
-  - `3:4` (portrait)
-  - `4:3` (landscape)
-  - `9:16` (vertical)
-  - `16:9` (wide)
-- **Features**: Photorealistic images with excellent quality
-
-### xAI Grok-2 Image
-
-- **Endpoint**: `https://api.x.ai/v1/images/generations`
-- **Model**: `grok-2-image-1212` (Grok's image generator)
-- **Features**: Text-to-image with creative interpretation
-- **Price**: $0.07 per image
-
-## Project Structure
-
-```
-image/
-├── SKILL.md              # Skill metadata with YAML frontmatter
-├── README.md             # This file
-├── package.json          # Bun package configuration
-├── tsconfig.json         # TypeScript configuration
-└── src/
-    ├── index.ts          # Main CLI entry point
-    ├── types.ts          # TypeScript type definitions
-    └── providers/
-        ├── openai.ts     # OpenAI DALL-E provider
-        ├── google.ts     # Google Imagen provider
-        └── xai.ts        # xAI Aurora provider
-```
-
-## Development
-
-The code is written in TypeScript and uses native Bun APIs:
-
-- `fetch` for HTTP requests (built into Bun)
-- File system operations via `fs/promises`
-- No external dependencies for image generation
-
-## Error Handling
-
-The skill includes comprehensive error handling:
-
-- API key validation
-- HTTP error responses
-- Invalid parameter detection
-- File system errors
-- Network failures
-
-## Examples
-
-### Generate a photorealistic image
-
-```bash
-bun run src/index.ts generate \
-  --provider google \
-  --prompt "a photorealistic portrait of a woman with red hair, studio lighting, professional photography" \
-  --output ./portrait.png
-```
-
-### Generate concept art
-
-```bash
-bun run src/index.ts generate \
-  --provider openai \
-  --prompt "concept art of a sci-fi spaceship, detailed, artstation trending" \
-  --output ./spaceship.png \
-  --size 1792x1024
-```
-
-### Generate creative interpretation
-
-```bash
-bun run src/index.ts generate \
-  --provider xai \
-  --prompt "the feeling of nostalgia as abstract art" \
-  --output ./nostalgia.png
-```
-
-## Help
-
-View help information:
+Poll and download results:
 
 ```bash
-bun run src/index.ts --help
+skills runs status <run-id>
+skills exports download <run-id>
 ```
 
-## License
+## Boundary
 
-MIT
+The OSS package contains metadata, pricing, and client contracts only. Provider
+credentials, model routing, prompts, moderation, billing, worker code, and
+artifact storage are owned by the hosted platform.

package/skills/webcrawling/CLAUDE.md

@@ -1,22 +1,11 @@
-# service-webcrawling
+# Web Crawling
 
-Web crawling service using Firecrawl API.
-
-## Tech Stack
-
-- Runtime: Bun
-- Language: TypeScript
-- CLI: Commander.js
-- API: Firecrawl
-
-## CLI
+Use this skill through the hosted Skills runtime.
 
 ```bash
-service-webcrawling scrape https://example.com
-service-webcrawling crawl https://example.com -d 2 -l 20
-service-webcrawling sessions
+skills run webcrawling --url https://example.com --depth 2
 ```
 
-## Environment
-
-- `FIRECRAWL_API_KEY` - Required
+The public package does not include provider credentials, crawler client code,
+hosted worker implementation, logs, or artifact storage. Authenticate with
+`skills auth login` before running.

package/skills/webcrawling/README.md

@@ -1,77 +1,24 @@
-# service-webcrawling
+# Web Crawling
 
-Web crawling service using Firecrawl API.
+Hosted web crawling and page extraction skill.
 
-## Installation
+## Usage
 
 ```bash
-bun install -g @hasnaxyz/service-webcrawling
+skills setup --mode hosted
+skills auth login
+skills run webcrawling --url https://example.com --depth 2
 ```
 
-## Quick Start
+Poll and download results:
 
 ```bash
-export FIRECRAWL_API_KEY=fc-...
-
-# Scrape a single URL
-service-webcrawling scrape https://example.com
-
-# Crawl a website
-service-webcrawling crawl https://example.com -d 2 -l 20
-
-# List sessions
-service-webcrawling sessions
-```
-
-## CLI Commands
-
-### scrape
-
-Scrape a single URL.
-
-```bash
-service-webcrawling scrape <url> [options]
-
-Options:
-  -f, --format <format>  markdown, html, json (default: markdown)
-  --full-page            Include full page, not just main content
-```
-
-### crawl
-
-Crawl an entire website.
-
-```bash
-service-webcrawling crawl <url> [options]
-
-Options:
-  -d, --depth <number>   Max depth (default: 2)
-  -l, --limit <number>   Max pages (default: 10)
-  --exclude <paths>      Exclude paths (comma-separated)
-  --include <paths>      Include only paths (comma-separated)
-```
-
-### sessions
-
-List crawl sessions.
-
-```bash
-service-webcrawling sessions [--json]
-```
-
-### config
-
-```bash
-service-webcrawling config view
-service-webcrawling config set firecrawlApiKey fc-...
-```
-
-## Environment Variables
-
-```bash
-FIRECRAWL_API_KEY=fc-...  # Required
+skills runs status <run-id>
+skills exports download <run-id>
 ```
 
-## License
+## Boundary
 
-MIT
+The OSS package contains metadata and documentation only. Crawling providers,
+credentials, rate limits, worker orchestration, logs, and artifacts are owned by
+the hosted runtime.

package/skills/browse/.env.example

@@ -1,14 +0,0 @@
-# Browser-Use API Configuration
-# Copy this file to .env and fill in your API key
-
-# Browser-Use Cloud API Key
-# Get your key from: https://cloud.browser-use.com/billing
-BROWSER_USE_API_KEY=bu_xxx...
-
-# Default LLM Model (optional)
-# Options: gpt-4o, o3, claude-sonnet-4, etc.
-# BROWSER_USE_MODEL=gpt-4o
-
-# Default proxy settings (optional)
-# BROWSER_USE_PROXY=true
-# BROWSER_USE_PROXY_COUNTRY=us

package/skills/deepresearch/.env.example

@@ -1,9 +0,0 @@
-# Required for search
-EXA_API_KEY=your_exa_api_key
-
-# Required for synthesis (at least one)
-ANTHROPIC_API_KEY=your_anthropic_api_key
-OPENAI_API_KEY=your_openai_api_key
-
-# Optional for deep scraping
-FIRECRAWL_API_KEY=your_firecrawl_api_key

package/skills/image/.env.example

@@ -1,12 +0,0 @@
-# OpenAI API Key
-# Get your key from: https://platform.openai.com/api-keys
-OPENAI_API_KEY=sk-...
-
-# Google Gemini API Key and Project ID
-# Get your key from: https://aistudio.google.com/apikey
-GEMINI_API_KEY=...
-GOOGLE_PROJECT_ID=your-project-id
-
-# xAI API Key
-# Get your key from: https://x.ai/api
-XAI_API_KEY=...

package/skills/image/PROJECT_OVERVIEW.md

@@ -1,216 +0,0 @@
-# Image Generation Skill - Project Overview
-
-## Summary
-
-A complete Claude Code skill for AI-powered image generation supporting three major providers: OpenAI DALL-E 3, Google Imagen 3, and xAI Aurora.
-
-## Key Statistics
-
-- **Total Lines of Code**: ~993 lines
-- **TypeScript Files**: 5 (457 lines)
-- **Providers**: 3 (OpenAI, Google, xAI)
-- **Documentation Files**: 5
-- **Example Scripts**: 3
-- **Dependencies**: Zero (uses native Bun APIs)
-
-## File Structure
-
-```
-image/
-├── SKILL.md              [56 lines]  - Skill metadata with YAML frontmatter
-├── README.md             [194 lines] - Comprehensive documentation
-├── QUICKSTART.md         [111 lines] - Quick start guide
-├── PROJECT_OVERVIEW.md   - This file
-├── package.json          [29 lines]  - Bun package configuration
-├── tsconfig.json         [17 lines]  - TypeScript configuration
-├── .env.example          [12 lines]  - Environment variable template
-├── .gitignore            [12 lines]  - Git ignore patterns
-├── examples/
-│   ├── openai-examples.sh    [35 lines] - OpenAI usage examples
-│   ├── google-examples.sh    [37 lines] - Google usage examples
-│   └── xai-examples.sh       [33 lines] - xAI usage examples
-└── src/
-    ├── index.ts          [189 lines] - Main CLI entry point
-    ├── types.ts          [37 lines]  - TypeScript type definitions
-    └── providers/
-        ├── openai.ts     [80 lines]  - OpenAI DALL-E provider
-        ├── google.ts     [78 lines]  - Google Imagen provider
-        └── xai.ts        [73 lines]  - xAI Aurora provider
-```
-
-## Features Implemented
-
-### Core Functionality
-- ✅ Multi-provider image generation (OpenAI, Google, xAI)
-- ✅ Clean CLI interface with argument parsing
-- ✅ Automatic image download and file saving
-- ✅ Support for custom models and sizes
-- ✅ Environment variable configuration
-- ✅ Comprehensive error handling
-
-### OpenAI Provider
-- ✅ DALL-E 3 support
-- ✅ GPT-4o image generation (gpt-image-1)
-- ✅ Three size options: 1024x1024, 1792x1024, 1024x1792
-- ✅ Revised prompt display
-- ✅ URL-based image retrieval
-
-### Google Imagen Provider
-- ✅ Imagen 3.0 support via Vertex AI REST API
-- ✅ Five aspect ratio options: 1:1, 3:4, 4:3, 9:16, 16:9
-- ✅ Base64 image decoding
-- ✅ Project ID configuration
-
-### xAI Aurora Provider
-- ✅ Aurora model support
-- ✅ OpenAI-compatible API format
-- ✅ Support for both URL and base64 responses
-- ✅ Flexible size configuration
-
-### Developer Experience
-- ✅ TypeScript with strict type checking
-- ✅ Native Bun runtime (no external dependencies)
-- ✅ Modular provider architecture
-- ✅ Clean separation of concerns
-- ✅ Comprehensive inline documentation
-- ✅ Example scripts for all providers
-
-### Error Handling
-- ✅ API key validation
-- ✅ Invalid provider detection
-- ✅ Missing argument detection
-- ✅ HTTP error handling
-- ✅ Network failure handling
-- ✅ Size validation (provider-specific)
-
-## Usage Examples
-
-### Basic Usage
-```bash
-bun run src/index.ts generate --provider openai --prompt "a cat" --output ./cat.png
-```
-
-### With Custom Options
-```bash
-bun run src/index.ts generate \
-  --provider openai \
-  --prompt "a futuristic city" \
-  --output ./city.png \
-  --size 1792x1024 \
-  --model dall-e-3
-```
-
-### Short Flags
-```bash
-bun run src/index.ts generate -p google --prompt "a dog" -o ./dog.png -s 16:9
-```
-
-## API Integration Details
-
-### OpenAI DALL-E 3
-- **Endpoint**: `https://api.openai.com/v1/images/generations`
-- **Authentication**: Bearer token via `Authorization` header
-- **Request Format**: JSON with model, prompt, size, response_format
-- **Response Format**: URL or base64
-- **Environment Variable**: `OPENAI_API_KEY`
-
-### Google Imagen 3
-- **Endpoint**: `https://{location}-aiplatform.googleapis.com/v1/projects/{projectId}/locations/{location}/publishers/google/models/{model}:predict`
-- **Authentication**: Bearer token via `Authorization` header
-- **Request Format**: JSON with instances and parameters
-- **Response Format**: Base64 encoded image
-- **Environment Variables**: `GOOGLE_API_KEY`, `GOOGLE_PROJECT_ID`
-
-### xAI Aurora
-- **Endpoint**: `https://api.x.ai/v1/images/generations`
-- **Authentication**: Bearer token via `Authorization` header
-- **Request Format**: OpenAI-compatible JSON
-- **Response Format**: URL or base64
-- **Environment Variable**: `XAI_API_KEY`
-
-## Architecture Patterns
-
-### Provider Interface
-All providers implement the `ImageProvider` interface:
-```typescript
-interface ImageProvider {
-  generate(prompt: string, options?: {
-    model?: string;
-    size?: string;
-  }): Promise<Buffer>;
-}
-```
-
-### Type Safety
-- Strong typing throughout the codebase
-- Provider-specific response interfaces
-- Validated command-line options
-- Type-safe error handling
-
-### Modularity
-- Each provider in separate file
-- Shared types in dedicated module
-- Clean CLI entry point
-- No circular dependencies
-
-## Testing Performed
-
-✅ Help command display
-✅ Invalid provider detection
-✅ Missing argument validation
-✅ Successful image generation (OpenAI)
-✅ File output and directory creation
-✅ TypeScript compilation
-✅ Bun runtime compatibility
-
-## Future Enhancement Possibilities
-
-- Image editing capabilities (inpainting, outpainting)
-- Batch generation support
-- Image variation generation
-- Progress bars for long-running operations
-- Local image storage caching
-- Configuration file support (JSON/YAML)
-- Additional providers (Stability AI, Midjourney API)
-- Image quality/resolution options
-- Rate limiting and retry logic
-- Cost estimation before generation
-
-## Technical Highlights
-
-1. **Zero Dependencies**: Uses only native Bun APIs (fetch, fs/promises)
-2. **Type-Safe**: Comprehensive TypeScript types for all operations
-3. **Clean Code**: Well-organized, readable, maintainable
-4. **Error Resilient**: Handles all common error scenarios gracefully
-5. **Extensible**: Easy to add new providers following existing pattern
-6. **Production Ready**: Includes all necessary error handling and validation
-
-## Quick Reference
-
-| Provider | Model | Default Size | Custom Sizes |
-|----------|-------|-------------|--------------|
-| OpenAI | dall-e-3 | 1024x1024 | 1792x1024, 1024x1792 |
-| Google | imagen-3.0-generate-001 | 1:1 | 3:4, 4:3, 9:16, 16:9 |
-| xAI | aurora | Default | Provider-specific |
-
-## Environment Setup
-
-```bash
-# OpenAI
-export OPENAI_API_KEY="sk-..."
-
-# Google
-export GOOGLE_API_KEY="..."
-export GOOGLE_PROJECT_ID="your-project"
-
-# xAI
-export XAI_API_KEY="..."
-```
-
-## License
-
-MIT
-
----
-
-Built with Bun, TypeScript, and Claude Code