@yeongjaeyou/claude-code-config 0.1.0

package/.claude/skills/code-explorer/references/github_api.md

@@ -0,0 +1,174 @@
+# GitHub CLI Reference
+
+## Installation
+
+```bash
+# macOS
+brew install gh
+
+# Ubuntu/Debian
+sudo apt install gh
+
+# Windows
+winget install GitHub.cli
+```
+
+## Authentication
+
+```bash
+# Login interactively
+gh auth login
+
+# Check status
+gh auth status
+```
+
+## Repository Search
+
+### Basic Search
+
+```bash
+# Search repositories
+gh search repos "keyword"
+
+# With options
+gh search repos "keyword" --limit 20 --sort stars
+
+# Filter by language
+gh search repos "keyword" --language python
+
+# Filter by owner
+gh search repos "keyword" --owner username
+```
+
+### Sort Options
+
+- `stars` - Sort by star count (default)
+- `forks` - Sort by fork count
+- `updated` - Sort by last update
+- `help-wanted-issues` - Sort by help-wanted issues
+- `best-match` - Sort by relevance
+
+### JSON Output
+
+```bash
+# Get specific fields as JSON
+gh search repos "keyword" --json fullName,description,stargazersCount,url
+
+# Available JSON fields
+# fullName, name, owner, description, url, createdAt, updatedAt
+# stargazersCount, forksCount, watchersCount, language, isArchived
+# isFork, isPrivate, visibility, license, topics
+```
+
+## Repository Operations
+
+### View Repository
+
+```bash
+# View in terminal
+gh repo view owner/repo
+
+# View specific fields as JSON
+gh repo view owner/repo --json name,description,stargazersCount,readme
+```
+
+### Clone Repository
+
+```bash
+# Clone
+gh repo clone owner/repo
+
+# Clone to specific directory
+gh repo clone owner/repo ./my-directory
+```
+
+### List Repository Contents
+
+```bash
+# List files (requires being in repo or using gh api)
+gh api repos/owner/repo/contents/path
+
+# Get file content
+gh api repos/owner/repo/contents/README.md --jq '.content' | base64 -d
+```
+
+## Code Search
+
+```bash
+# Search code
+gh search code "function_name"
+
+# Filter by file extension
+gh search code "import torch" --extension py
+
+# Filter by repository
+gh search code "keyword" --repo owner/repo
+```
+
+## API Usage
+
+### Direct API Calls
+
+```bash
+# GET request
+gh api repos/owner/repo
+
+# With pagination
+gh api repos/owner/repo/commits --paginate
+
+# With query parameters
+gh api search/repositories -f q="keyword" -f sort=stars
+```
+
+### Common Endpoints
+
+```bash
+# Repository info
+gh api repos/owner/repo
+
+# Repository contents
+gh api repos/owner/repo/contents/path
+
+# Releases
+gh api repos/owner/repo/releases
+
+# Contributors
+gh api repos/owner/repo/contributors
+```
+
+## Useful Examples
+
+### Find Popular Python ML Repositories
+
+```bash
+gh search repos "machine learning" --language python --sort stars --limit 20
+```
+
+### Get Repository Structure
+
+```bash
+gh api repos/owner/repo/git/trees/main?recursive=1 --jq '.tree[].path'
+```
+
+### Download Specific File
+
+```bash
+curl -sL "https://raw.githubusercontent.com/owner/repo/main/path/to/file"
+```
+
+### Check Repository Activity
+
+```bash
+gh api repos/owner/repo/commits --jq '.[0:5] | .[] | "\(.commit.author.date) - \(.commit.message | split("\n")[0])"'
+```
+
+## Rate Limits
+
+- Authenticated: 5,000 requests/hour
+- Search API: 30 requests/minute
+
+Check rate limit:
+```bash
+gh api rate_limit
+```

package/.claude/skills/code-explorer/references/huggingface_api.md

@@ -0,0 +1,240 @@
+# Hugging Face Hub Reference
+
+## Installation
+
+```bash
+# For uvx hf (no install needed)
+uvx hf --help
+
+# Or install huggingface_hub globally
+pip install -U huggingface_hub
+```
+
+## Authentication
+
+```bash
+# Login interactively
+uvx hf auth login
+
+# Or set token
+export HF_TOKEN="your_token_here"
+```
+
+---
+
+# uvx hf CLI Commands
+
+Note: `hf` CLI does NOT have search commands. Use Python API for searching.
+
+## Download Commands
+
+### Download Repository
+
+```bash
+# Download entire model
+uvx hf download <repo_id>
+
+# Download specific files
+uvx hf download <repo_id> --include "*.py"
+uvx hf download <repo_id> --include "config.json" --include "*.safetensors"
+
+# Download to specific directory
+uvx hf download <repo_id> --local-dir ./my-model
+
+# Download from specific repo type
+uvx hf download <repo_id> --repo-type space
+uvx hf download <repo_id> --repo-type dataset
+```
+
+### Download Examples
+
+```bash
+# Download model config (temporary analysis)
+uvx hf download microsoft/resnet-50 --include "*.json" --local-dir /tmp/resnet-50
+
+# Download space source code (temporary analysis)
+uvx hf download Qwen/Qwen3-VL-Demo --repo-type space --include "*.py" --local-dir /tmp/qwen3-vl-demo
+
+# Download specific file
+uvx hf download Qwen/Qwen2.5-VL-7B-Instruct config.json --local-dir /tmp/qwen-config
+```
+
+**Note**: Always use `--local-dir /tmp/` for temporary code analysis to avoid cluttering the project.
+
+## Repository Commands
+
+```bash
+# View repo info
+uvx hf repo info <repo_id>
+
+# List files in repo
+uvx hf repo-files list <repo_id>
+
+# Create new repo
+uvx hf repo create <repo_name>
+uvx hf repo create <repo_name> --type space
+```
+
+## Upload Commands
+
+```bash
+# Upload file
+uvx hf upload <repo_id> <local_path> <path_in_repo>
+
+# Upload folder
+uvx hf upload <repo_id> ./local_folder
+```
+
+---
+
+# Python API (huggingface_hub) - For Searching
+
+## Basic Setup
+
+```python
+from huggingface_hub import HfApi
+
+api = HfApi()
+```
+
+## Search Models
+
+```python
+# Basic search
+models = list(api.list_models(search="object detection", limit=10))
+
+# With filters
+models = list(api.list_models(
+    search="qwen vl",
+    task="image-text-to-text",
+    sort="downloads",
+    direction=-1,
+    limit=20
+))
+
+# Print results
+for m in models:
+    print(f"{m.id}: {m.downloads:,} downloads")
+```
+
+### Model Tasks
+
+Common task filters:
+- `object-detection`
+- `image-segmentation`
+- `image-classification`
+- `image-text-to-text`
+- `text-generation`
+
+## Search Datasets
+
+```python
+datasets = list(api.list_datasets(search="coco", limit=10))
+
+for d in datasets:
+    print(f"{d.id}: {d.downloads:,} downloads")
+```
+
+## Search Spaces
+
+```python
+spaces = list(api.list_spaces(search="gradio demo", limit=10))
+
+# Filter by SDK
+spaces = list(api.list_spaces(
+    search="object detection",
+    sdk="gradio",
+    sort="likes",
+    direction=-1,
+    limit=20
+))
+
+for s in spaces:
+    print(f"{s.id}: {s.likes} likes, SDK: {s.sdk}")
+```
+
+## Download Files (Python)
+
+```python
+from huggingface_hub import hf_hub_download
+
+# Download model file
+file_path = hf_hub_download(
+    repo_id="microsoft/resnet-50",
+    filename="config.json"
+)
+
+# Download from space
+file_path = hf_hub_download(
+    repo_id="Qwen/Qwen3-VL-Demo",
+    filename="app.py",
+    repo_type="space"
+)
+```
+
+## List Repository Files
+
+```python
+# List model files
+files = api.list_repo_files("microsoft/resnet-50")
+
+# List space files
+files = api.list_repo_files("Qwen/Qwen3-VL-Demo", repo_type="space")
+```
+
+---
+
+# Web URLs
+
+## Direct Access
+
+```
+Models:   https://huggingface.co/{model_id}
+Datasets: https://huggingface.co/datasets/{dataset_id}
+Spaces:   https://huggingface.co/spaces/{space_id}
+```
+
+## View Files
+
+```
+https://huggingface.co/{model_id}/tree/main
+https://huggingface.co/spaces/{space_id}/tree/main
+```
+
+## Raw File Access
+
+```
+https://huggingface.co/{model_id}/raw/main/{filename}
+https://huggingface.co/spaces/{space_id}/raw/main/app.py
+```
+
+---
+
+# Common Patterns
+
+## Search + Download Workflow
+
+```bash
+# 1. Search for spaces (using script)
+python scripts/search_huggingface.py "qwen vl" --type spaces
+
+# 2. Download the source code (to /tmp/ for temporary analysis)
+uvx hf download Qwen/Qwen3-VL-Demo --repo-type space --include "*.py" --local-dir /tmp/qwen-demo
+```
+
+## Analyze Space Implementation
+
+```bash
+# Download app.py and requirements (to /tmp/)
+uvx hf download username/space-name --repo-type space --include "app.py" --include "requirements.txt" --local-dir /tmp/space-code
+
+# Read the code
+cat /tmp/space-code/app.py
+```
+
+---
+
+# Rate Limits
+
+- Anonymous: 1,000 requests/hour
+- Authenticated: 10,000 requests/hour

package/.claude/skills/code-explorer/scripts/search_github.py

@@ -0,0 +1,208 @@
+#!/usr/bin/env python3
+"""
+GitHub Repository Search Script
+
+Search GitHub repositories using the gh CLI.
+Requires: gh CLI installed and authenticated (https://cli.github.com/)
+
+Usage:
+    python search_github.py "keyword" [options]
+
+Examples:
+    python search_github.py "object detection" --limit 10
+    python search_github.py "gradio app" --language python
+    python search_github.py "yolo world" --detailed --sort stars
+"""
+
+import argparse
+import json
+import subprocess
+import sys
+from typing import Optional
+
+
+def check_gh_cli() -> bool:
+    """Check if gh CLI is installed and authenticated."""
+    try:
+        result = subprocess.run(
+            ["gh", "auth", "status"],
+            capture_output=True,
+            text=True
+        )
+        return result.returncode == 0
+    except FileNotFoundError:
+        return False
+
+
+def search_repos(
+    query: str,
+    limit: int = 10,
+    language: Optional[str] = None,
+    sort: str = "stars",
+    detailed: bool = False
+) -> list[dict]:
+    """
+    Search GitHub repositories.
+
+    Args:
+        query: Search keyword
+        limit: Maximum number of results
+        language: Filter by programming language
+        sort: Sort by 'stars', 'forks', or 'updated'
+        detailed: Include detailed info (description, topics)
+
+    Returns:
+        List of repository information dictionaries
+    """
+    cmd = [
+        "gh", "search", "repos", query,
+        "--limit", str(limit),
+        "--sort", sort,
+        "--json", "fullName,description,stargazersCount,forksCount,updatedAt,language,url,isArchived"
+    ]
+
+    if language:
+        cmd.extend(["--language", language])
+
+    try:
+        result = subprocess.run(cmd, capture_output=True, text=True, check=True)
+        repos = json.loads(result.stdout)
+        return repos
+    except subprocess.CalledProcessError as e:
+        print(f"Error searching repositories: {e.stderr}", file=sys.stderr)
+        return []
+    except json.JSONDecodeError as e:
+        print(f"Error parsing response: {e}", file=sys.stderr)
+        return []
+
+
+def get_repo_details(repo_name: str) -> Optional[dict]:
+    """
+    Get detailed information about a repository.
+
+    Args:
+        repo_name: Repository full name (owner/repo)
+
+    Returns:
+        Repository details dictionary or None
+    """
+    cmd = [
+        "gh", "repo", "view", repo_name,
+        "--json", "name,owner,description,stargazersCount,forksCount,watchers,issues,pullRequests,url,homepageUrl,createdAt,updatedAt,pushedAt,isArchived,isFork,languages,repositoryTopics,licenseInfo,readme"
+    ]
+
+    try:
+        result = subprocess.run(cmd, capture_output=True, text=True, check=True)
+        return json.loads(result.stdout)
+    except subprocess.CalledProcessError as e:
+        print(f"Error fetching repo details: {e.stderr}", file=sys.stderr)
+        return None
+    except json.JSONDecodeError as e:
+        print(f"Error parsing response: {e}", file=sys.stderr)
+        return None
+
+
+def format_repo_summary(repo: dict) -> str:
+    """Format repository info as a summary string."""
+    archived = " [ARCHIVED]" if repo.get("isArchived") else ""
+    lang = repo.get("language") or "N/A"
+
+    return (
+        f"\n{'='*60}\n"
+        f"{repo['fullName']}{archived}\n"
+        f"{'='*60}\n"
+        f"Stars: {repo['stargazersCount']:,} | Forks: {repo['forksCount']:,} | Language: {lang}\n"
+        f"Updated: {repo['updatedAt'][:10]}\n"
+        f"URL: {repo['url']}\n"
+        f"\nDescription:\n{repo.get('description') or 'No description'}\n"
+    )
+
+
+def format_repo_detailed(details: dict) -> str:
+    """Format detailed repository info."""
+    topics = details.get("repositoryTopics", [])
+    topic_names = [t.get("name", "") for t in topics] if topics else []
+
+    languages = details.get("languages", [])
+    lang_names = [l.get("node", {}).get("name", "") for l in languages] if languages else []
+
+    license_info = details.get("licenseInfo")
+    license_name = license_info.get("name", "N/A") if license_info else "N/A"
+
+    readme = details.get("readme") or "No README available"
+    # Truncate README if too long
+    if len(readme) > 2000:
+        readme = readme[:2000] + "\n... [truncated]"
+
+    return (
+        f"\n--- Detailed Info ---\n"
+        f"Topics: {', '.join(topic_names) if topic_names else 'None'}\n"
+        f"Languages: {', '.join(lang_names) if lang_names else 'N/A'}\n"
+        f"License: {license_name}\n"
+        f"Created: {details.get('createdAt', 'N/A')[:10]}\n"
+        f"Last Push: {details.get('pushedAt', 'N/A')[:10]}\n"
+        f"Open Issues: {details.get('issues', {}).get('totalCount', 'N/A')}\n"
+        f"Open PRs: {details.get('pullRequests', {}).get('totalCount', 'N/A')}\n"
+        f"\n--- README ---\n{readme}\n"
+    )
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Search GitHub repositories",
+        formatter_class=argparse.RawDescriptionHelpFormatter,
+        epilog="""
+Examples:
+  %(prog)s "object detection" --limit 10
+  %(prog)s "gradio app" --language python
+  %(prog)s "yolo world" --detailed --sort stars
+        """
+    )
+
+    parser.add_argument("query", help="Search keyword")
+    parser.add_argument("--limit", "-n", type=int, default=10, help="Number of results (default: 10)")
+    parser.add_argument("--language", "-l", help="Filter by programming language")
+    parser.add_argument("--sort", "-s", choices=["stars", "forks", "updated"], default="stars", help="Sort criteria (default: stars)")
+    parser.add_argument("--detailed", "-d", action="store_true", help="Show detailed info including README")
+    parser.add_argument("--json", "-j", action="store_true", help="Output as JSON")
+
+    args = parser.parse_args()
+
+    # Check gh CLI
+    if not check_gh_cli():
+        print("Error: gh CLI is not installed or not authenticated.", file=sys.stderr)
+        print("Install: https://cli.github.com/", file=sys.stderr)
+        print("Then run: gh auth login", file=sys.stderr)
+        sys.exit(1)
+
+    # Search repositories
+    print(f"Searching GitHub for: '{args.query}'...\n", file=sys.stderr)
+    repos = search_repos(
+        query=args.query,
+        limit=args.limit,
+        language=args.language,
+        sort=args.sort,
+        detailed=args.detailed
+    )
+
+    if not repos:
+        print("No repositories found.", file=sys.stderr)
+        sys.exit(0)
+
+    # Output results
+    if args.json:
+        print(json.dumps(repos, indent=2))
+    else:
+        print(f"Found {len(repos)} repositories:\n")
+
+        for i, repo in enumerate(repos, 1):
+            print(f"[{i}] {format_repo_summary(repo)}")
+
+            if args.detailed:
+                details = get_repo_details(repo["fullName"])
+                if details:
+                    print(format_repo_detailed(details))
+
+
+if __name__ == "__main__":
+    main()