agent-recipes 0.0.5__py3-none-any.whl

agent_recipes/templates/ai-color-palette-extractor/README.md

@@ -0,0 +1,60 @@
+# Color Palette Extractor
+
+Extract dominant colors from images
+
+## Installation
+
+```bash
+pip install agent-recipes
+```
+
+## CLI Usage
+
+```bash
+# Basic usage
+praison recipes run ai-color-palette-extractor <input>
+
+# With output directory
+praison recipes run ai-color-palette-extractor <input> --output ./output/
+
+# Dry run (check dependencies only)
+praison recipes run ai-color-palette-extractor <input> --dry-run
+
+# Show recipe info
+praison recipes info ai-color-palette-extractor
+
+# Check dependencies
+praison recipes doctor ai-color-palette-extractor
+```
+
+## Python Usage
+
+```python
+from praisonaiagents import Agent
+from agent_recipes import load_recipe
+
+# Load and run recipe
+recipe = load_recipe("ai-color-palette-extractor")
+result = recipe.run(input="<your-input>")
+print(result)
+```
+
+## Requirements
+
+### Packages
+- pillow
+- colorthief
+
+### Environment Variables
+- None
+
+### External Tools
+- None
+
+## Tags
+
+`image`, `color`, `palette`
+
+## License
+
+Apache-2.0

agent_recipes/templates/ai-color-palette-extractor/TEMPLATE.yaml

@@ -0,0 +1,27 @@
+name: ai-color-palette-extractor
+version: "1.0.0"
+description: Extract dominant colors from images
+author: praison
+license: Apache-2.0
+tags:
+  - image
+  - color
+  - palette
+
+requires:
+  tools:
+    - image_tool
+  packages:
+    - pillow
+    - colorthief
+
+cli:
+  command: praison recipes run ai-color-palette-extractor
+  examples:
+    - praison recipes run ai-color-palette-extractor <input>
+    - praison recipes run ai-color-palette-extractor <input> --output ./output/
+    - praison recipes run ai-color-palette-extractor <input> --dry-run
+
+safety:
+  dry_run_default: false
+  overwrites_files: true

agent_recipes/templates/ai-comment-miner/recipe.yaml

@@ -0,0 +1,40 @@
+name: ai-comment-miner
+version: 1.0.0
+description: Mine comments for content ideas and audience insights
+author: PraisonAI
+tags:
+  - comments
+  - ideas
+  - audience
+  - mining
+
+config:
+  extract_ideas: true
+  sentiment_analysis: true
+  max_comments: 100
+
+input:
+  type: object
+  properties:
+    comments:
+      type: array
+
+output:
+  type: object
+  properties:
+    ideas:
+      type: array
+    insights:
+      type: object
+
+requires:
+  env:
+    - OPENAI_API_KEY
+
+workflow:
+  agents:
+    - name: miner
+      role: Comment Miner
+      tools:
+        - extract_ideas
+        - analyze_sentiment

agent_recipes/templates/ai-comment-miner/tools.py

@@ -0,0 +1,141 @@
+"""
+AI Comment Miner Tools
+
+Extract content ideas and insights from comments.
+"""
+
+import logging
+import os
+from typing import Any, Dict, List
+
+logger = logging.getLogger(__name__)
+
+
+def call_llm(prompt: str, max_tokens: int = 800) -> str:
+    """Call OpenAI API."""
+    import requests
+    api_key = os.environ.get("OPENAI_API_KEY")
+    if not api_key:
+        raise ValueError("OPENAI_API_KEY not set")
+    response = requests.post(
+        "https://api.openai.com/v1/chat/completions",
+        headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
+        json={"model": "gpt-4o-mini", "messages": [{"role": "user", "content": prompt}], "max_tokens": max_tokens},
+        timeout=60,
+    )
+    response.raise_for_status()
+    return response.json()["choices"][0]["message"]["content"]
+
+
+def extract_ideas(
+    comments: List[str],
+    max_ideas: int = 10,
+) -> Dict[str, Any]:
+    """
+    Extract content ideas from comments.
+    
+    Args:
+        comments: List of comment texts
+        max_ideas: Maximum ideas to extract
+        
+    Returns:
+        Dictionary with extracted ideas
+    """
+    comments_text = "\n".join([f"- {c[:200]}" for c in comments[:50]])
+    
+    prompt = f"""Analyze these audience comments and extract content ideas:
+
+{comments_text}
+
+Extract:
+1. Questions people are asking (potential video topics)
+2. Pain points mentioned (problems to solve)
+3. Requests for content (what they want to see)
+4. Trending topics mentioned
+
+Provide {max_ideas} specific, actionable content ideas based on these comments.
+Format each idea on its own line."""
+    
+    try:
+        result = call_llm(prompt, max_tokens=600)
+        
+        ideas = []
+        for line in result.split("\n"):
+            line = line.strip().lstrip("-•0123456789.").strip()
+            if line and len(line) > 10:
+                ideas.append(line)
+        
+        return {
+            "ideas": ideas[:max_ideas],
+            "comments_analyzed": len(comments),
+        }
+    except Exception as e:
+        logger.error(f"Error extracting ideas: {e}")
+        return {"error": str(e)}
+
+
+def analyze_sentiment(
+    comments: List[str],
+) -> Dict[str, Any]:
+    """
+    Analyze sentiment of comments.
+    
+    Args:
+        comments: List of comment texts
+        
+    Returns:
+        Sentiment analysis results
+    """
+    positive = 0
+    negative = 0
+    neutral = 0
+    
+    positive_words = ["love", "great", "awesome", "amazing", "helpful", "thanks", "best", "excellent"]
+    negative_words = ["hate", "bad", "terrible", "worst", "boring", "waste", "disappointed"]
+    
+    for comment in comments:
+        comment_lower = comment.lower()
+        pos_count = sum(1 for w in positive_words if w in comment_lower)
+        neg_count = sum(1 for w in negative_words if w in comment_lower)
+        
+        if pos_count > neg_count:
+            positive += 1
+        elif neg_count > pos_count:
+            negative += 1
+        else:
+            neutral += 1
+    
+    total = len(comments)
+    
+    return {
+        "sentiment": {
+            "positive": positive,
+            "negative": negative,
+            "neutral": neutral,
+            "positive_rate": round(positive / total * 100, 1) if total > 0 else 0,
+            "negative_rate": round(negative / total * 100, 1) if total > 0 else 0,
+        },
+        "total_analyzed": total,
+    }
+
+
+def mine_comments(
+    comments: List[str],
+) -> Dict[str, Any]:
+    """
+    Full comment mining pipeline.
+    
+    Args:
+        comments: List of comment texts
+        
+    Returns:
+        Complete mining results
+    """
+    ideas = extract_ideas(comments)
+    sentiment = analyze_sentiment(comments)
+    
+    return {
+        "ideas": ideas.get("ideas", []),
+        "sentiment": sentiment.get("sentiment", {}),
+        "total_comments": len(comments),
+    }

agent_recipes/templates/ai-commit-message-generator/README.md

@@ -0,0 +1,59 @@
+# Commit Message Generator
+
+Generate git commit messages from diffs
+
+## Installation
+
+```bash
+pip install agent-recipes
+```
+
+## CLI Usage
+
+```bash
+# Basic usage
+praison recipes run ai-commit-message-generator <input>
+
+# With output directory
+praison recipes run ai-commit-message-generator <input> --output ./output/
+
+# Dry run (check dependencies only)
+praison recipes run ai-commit-message-generator <input> --dry-run
+
+# Show recipe info
+praison recipes info ai-commit-message-generator
+
+# Check dependencies
+praison recipes doctor ai-commit-message-generator
+```
+
+## Python Usage
+
+```python
+from praisonaiagents import Agent
+from agent_recipes import load_recipe
+
+# Load and run recipe
+recipe = load_recipe("ai-commit-message-generator")
+result = recipe.run(input="<your-input>")
+print(result)
+```
+
+## Requirements
+
+### Packages
+- openai
+
+### Environment Variables
+- OPENAI_API_KEY
+
+### External Tools
+- git
+
+## Tags
+
+`code`, `git`, `commit`
+
+## License
+
+Apache-2.0

agent_recipes/templates/ai-commit-message-generator/TEMPLATE.yaml

@@ -0,0 +1,31 @@
+name: ai-commit-message-generator
+version: "1.0.0"
+description: Generate git commit messages from diffs
+author: praison
+license: Apache-2.0
+tags:
+  - code
+  - git
+  - commit
+
+requires:
+  tools:
+    - repo_tool
+    - llm_tool
+  packages:
+    - openai
+  env:
+    - OPENAI_API_KEY
+  external:
+    - git
+
+cli:
+  command: praison recipes run ai-commit-message-generator
+  examples:
+    - praison recipes run ai-commit-message-generator <input>
+    - praison recipes run ai-commit-message-generator <input> --output ./output/
+    - praison recipes run ai-commit-message-generator <input> --dry-run
+
+safety:
+  dry_run_default: true
+  overwrites_files: true

agent_recipes/templates/ai-content-calendar/recipe.yaml

@@ -0,0 +1,43 @@
+name: ai-content-calendar
+version: 1.0.0
+description: Generate content calendars with what/when/where scheduling
+author: PraisonAI
+tags:
+  - calendar
+  - scheduling
+  - planning
+  - content
+
+config:
+  schedule_frequency: daily
+  platforms:
+    - youtube
+    - x
+    - linkedin
+  timezone: UTC
+
+input:
+  type: object
+  properties:
+    topics:
+      type: array
+    duration_days:
+      type: integer
+
+output:
+  type: object
+  properties:
+    calendar:
+      type: array
+
+requires:
+  env:
+    - OPENAI_API_KEY
+
+workflow:
+  agents:
+    - name: calendar_planner
+      role: Content Calendar Planner
+      tools:
+        - generate_calendar
+        - optimize_schedule

agent_recipes/templates/ai-content-calendar/tools.py

@@ -0,0 +1,170 @@
+"""
+AI Content Calendar Tools
+
+Generate content calendars with:
+- Optimal posting times
+- Platform-specific scheduling
+- Topic distribution
+"""
+
+import json
+import logging
+import os
+from datetime import datetime, timedelta, timezone
+from typing import Any, Dict, List, Optional
+
+logger = logging.getLogger(__name__)
+
+
+def call_llm(prompt: str, max_tokens: int = 1000) -> str:
+    """Call OpenAI API."""
+    import requests
+    
+    api_key = os.environ.get("OPENAI_API_KEY")
+    if not api_key:
+        raise ValueError("OPENAI_API_KEY not set")
+    
+    response = requests.post(
+        "https://api.openai.com/v1/chat/completions",
+        headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
+        json={"model": "gpt-4o-mini", "messages": [{"role": "user", "content": prompt}], "max_tokens": max_tokens},
+        timeout=60,
+    )
+    response.raise_for_status()
+    return response.json()["choices"][0]["message"]["content"]
+
+
+def generate_calendar(
+    topics: List[str],
+    duration_days: int = 30,
+    platforms: Optional[List[str]] = None,
+    posts_per_day: int = 1,
+) -> Dict[str, Any]:
+    """
+    Generate a content calendar.
+    
+    Args:
+        topics: List of content topics
+        duration_days: Calendar duration in days
+        platforms: Target platforms
+        posts_per_day: Posts per day
+        
+    Returns:
+        Dictionary with calendar entries
+    """
+    platforms = platforms or ["youtube", "x", "linkedin"]
+    
+    # Optimal posting times by platform
+    optimal_times = {
+        "youtube": ["09:00", "12:00", "17:00"],
+        "x": ["08:00", "12:00", "17:00", "20:00"],
+        "linkedin": ["07:30", "12:00", "17:30"],
+        "instagram": ["11:00", "14:00", "19:00"],
+        "tiktok": ["07:00", "12:00", "19:00", "22:00"],
+    }
+    
+    calendar = []
+    start_date = datetime.now(timezone.utc).replace(hour=0, minute=0, second=0, microsecond=0)
+    
+    topic_idx = 0
+    
+    for day in range(duration_days):
+        current_date = start_date + timedelta(days=day)
+        
+        for post_num in range(posts_per_day):
+            # Rotate through topics
+            topic = topics[topic_idx % len(topics)]
+            topic_idx += 1
+            
+            # Rotate through platforms
+            platform = platforms[post_num % len(platforms)]
+            
+            # Get optimal time
+            times = optimal_times.get(platform, ["12:00"])
+            post_time = times[post_num % len(times)]
+            
+            calendar.append({
+                "date": current_date.strftime("%Y-%m-%d"),
+                "time": post_time,
+                "platform": platform,
+                "topic": topic,
+                "status": "scheduled",
+            })
+    
+    return {
+        "calendar": calendar,
+        "stats": {
+            "total_posts": len(calendar),
+            "duration_days": duration_days,
+            "platforms": platforms,
+        }
+    }
+
+
+def optimize_schedule(
+    calendar: List[Dict[str, Any]],
+    avoid_weekends: bool = False,
+    peak_hours_only: bool = True,
+) -> Dict[str, Any]:
+    """
+    Optimize a content calendar schedule.
+    
+    Args:
+        calendar: Existing calendar entries
+        avoid_weekends: Skip weekend posts
+        peak_hours_only: Only schedule during peak hours
+        
+    Returns:
+        Optimized calendar
+    """
+    optimized = []
+    
+    for entry in calendar:
+        date_str = entry.get("date", "")
+        
+        try:
+            date = datetime.strptime(date_str, "%Y-%m-%d")
+            
+            # Skip weekends if requested
+            if avoid_weekends and date.weekday() >= 5:
+                continue
+            
+            optimized.append(entry)
+            
+        except ValueError:
+            optimized.append(entry)
+    
+    return {
+        "calendar": optimized,
+        "removed": len(calendar) - len(optimized),
+    }
+
+
+def export_calendar(
+    calendar: List[Dict[str, Any]],
+    output_path: str,
+    format: str = "json",
+) -> Dict[str, Any]:
+    """
+    Export calendar to file.
+    
+    Args:
+        calendar: Calendar entries
+        output_path: Output file path
+        format: Export format (json, csv, ical)
+        
+    Returns:
+        Export info
+    """
+    if format == "json":
+        with open(output_path, "w") as f:
+            json.dump(calendar, f, indent=2)
+    elif format == "csv":
+        import csv
+        with open(output_path, "w", newline="") as f:
+            if calendar:
+                writer = csv.DictWriter(f, fieldnames=calendar[0].keys())
+                writer.writeheader()
+                writer.writerows(calendar)
+    
+    return {"path": output_path, "format": format, "entries": len(calendar)}

agent_recipes/templates/ai-context-enricher/recipe.yaml

@@ -0,0 +1,48 @@
+name: ai-context-enricher
+version: 1.0.0
+description: Enrich news articles with background context, prior art, stakeholder analysis, and hype detection
+author: PraisonAI
+tags:
+  - news
+  - enrichment
+  - context
+  - analysis
+
+config:
+  enrichment_types:
+    - background
+    - prior_art
+    - stakeholders
+    - hype_detection
+  max_context_length: 500
+  output_format: json
+
+input:
+  type: object
+  properties:
+    articles:
+      type: array
+      description: Articles to enrich
+
+output:
+  type: object
+  properties:
+    enriched_articles:
+      type: array
+
+requires:
+  env:
+    - OPENAI_API_KEY
+  optional_env:
+    - TAVILY_API_KEY
+
+workflow:
+  agents:
+    - name: enricher
+      role: Context Analyst
+      goal: Add rich context to news articles
+      tools:
+        - add_background
+        - find_prior_art
+        - analyze_stakeholders
+        - detect_hype