Skip to main content

🎯 Easy Module

The easy/ module provides one-liner APIs for common AI tasks. Zero configuration, instant results.

Module Overview​

File Structure​

src/openstackai/easy/
├── __init__.py
├── ask.py              # Universal Q&A
├── research.py         # Deep research
├── summarize.py        # Summarization
├── extract.py          # Data extraction
├── generate.py         # Content generation
├── translate.py        # Translation
├── fetch.py            # Real-time data
├── analyze.py          # Data analysis
├── rag.py              # RAG system
├── code.py             # Code operations
├── chat.py             # Interactive chat
├── handoff.py          # Agent handoffs
├── guardrails.py       # Safety guards
├── trace.py            # Tracing/logging
├── mcp.py              # MCP protocol
├── agent_factory.py    # Quick agent creation
├── config.py           # Auto-configuration
└── llm_interface.py    # LLM abstraction

Core Functions​

ask() — Universal Question Answering​

from openstackai import ask

# Simple question
answer = ask("What is Python?")

# With options
answer = ask(
    "Explain quantum computing",
    detailed=True,        # Longer response
    format="bullet",      # Bullet points
    model="gpt-4"         # Specific model
)

Parameters:

ParameterTypeDefaultDescription
questionstrrequiredThe question
detailedboolFalseLonger response
formatstrNone"bullet", "numbered", "json"
modelstrautoModel override

research() — Deep Topic Research​

from openstackai import research

result = research("AI trends in enterprise software")

print(result.summary)      # Executive summary
print(result.key_points)   # List of key findings
print(result.insights)     # Strategic insights
print(result.sources)      # Referenced sources

Returns: ResearchResult

FieldTypeDescription
summarystrExecutive summary
key_pointslistKey findings
insightslistStrategic insights
sourceslistSource references

summarize() — Summarization​

from openstackai import summarize

# Summarize text
summary = summarize(long_text)

# Summarize a file
summary = summarize("./report.pdf")
summary = summarize("./document.docx")

# Summarize a URL
summary = summarize("https://example.com/article")

# With length control
summary = summarize(text, length="brief")      # ~50 words
summary = summarize(text, length="detailed")   # ~300 words

extract() — Structured Data Extraction​

from openstackai import extract

# Extract specific fields
data = extract(
    invoice_text,
    fields=["invoice_number", "date", "total", "items"]
)

# Returns structured data
print(data["invoice_number"])  # "INV-2024-001"
print(data["total"])           # "1,234.56"
print(data["items"])           # [{"name": "Widget", "qty": 5}]

# With schema
from pydantic import BaseModel

class Person(BaseModel):
    name: str
    age: int
    city: str

person = extract(text, schema=Person)

generate() — Content Generation​

from openstackai import generate

# Generate different content types
blog = generate("Python async programming", type="article")
email = generate("thank you letter", type="email")
code = generate("fibonacci function", type="code")
sql = generate("users table with auth", type="sql")

# With customization
content = generate(
    "API documentation",
    type="markdown",
    style="technical",
    length="detailed"
)

translate() — Translation​

from openstackai import translate

# Simple translation
spanish = translate("Hello, how are you?", to="spanish")
japanese = translate("Hello", to="japanese")

# Auto-detect source language
result = translate("Bonjour le monde", to="english")

Data Functions​

fetch — Real-time Data​

from openstackai import fetch

# Weather
weather = fetch.weather("New York")
print(weather.temperature)
print(weather.conditions)

# News
news = fetch.news("artificial intelligence")
for article in news.articles:
    print(article.title)
    print(article.summary)

# Stock data
stock = fetch.stock("AAPL")
print(stock.price)
print(stock.change)

# Any URL
content = fetch.url("https://api.example.com/data")

analyze — Data Analysis​

from openstackai import analyze

# Sentiment analysis
sentiment = analyze.sentiment("I love this product!")
print(sentiment.score)      # 0.9
print(sentiment.label)      # "positive"

# Entity extraction
entities = analyze.entities(text)
for entity in entities:
    print(entity.text, entity.type)

# Topic classification  
topics = analyze.topics(document)

rag — RAG System​

from openstackai import rag

# Index documents
docs = rag.index("./documents/")
docs = rag.index(["doc1.pdf", "doc2.md"])

# Ask questions
answer = docs.ask("What is the main conclusion?")
answer = docs.ask("Summarize the findings")

# With sources
result = docs.ask("What are the recommendations?", return_sources=True)
print(result.answer)
print(result.sources)

RAG Architecture:

Code Functions​

code — Code Operations​

from openstackai import code

# Write new code
api = code.write("REST API with CRUD operations")
function = code.write("async file downloader")

# Review existing code
review = code.review(my_code)
print(review.issues)         # List of issues
print(review.suggestions)    # Improvement suggestions
print(review.score)          # Quality score

# Debug errors
fix = code.debug("""
TypeError: 'NoneType' object is not subscriptable
At line 42: result = data['key']
""")
print(fix.explanation)
print(fix.solution)

# Refactor code
improved = code.refactor(
    old_code,
    goal="convert to async"
)

# Generate tests
tests = code.test(my_function)

Advanced Functions​

handoff — Agent Handoffs​

from openstackai import handoff, Agent

agent_a = Agent(name="Researcher")
agent_b = Agent(name="Writer")

# Transfer control with context
result = handoff(
    from_agent=agent_a,
    to_agent=agent_b,
    context={"research": research_data}
)

guardrails — Safety Guards​

from openstackai import guardrails

# Create guarded function
safe_ask = guardrails.wrap(
    ask,
    block_pii=True,           # Block PII
    block_harmful=True,       # Block harmful content
    max_tokens=1000,          # Limit response
    allowed_topics=["tech"]   # Topic filter
)

answer = safe_ask("Tell me about Python")

trace — Tracing​

from openstackai import trace

# Enable tracing
trace.enable()

# Run operations (automatically traced)
answer = ask("What is AI?")

# View traces
trace.show()

# Export traces
trace.export("traces.json")

chat — Interactive Sessions​

from openstackai import chat

# Start a chat session
session = chat.start()

# Conversation with memory
response1 = session.send("My name is Alice")
response2 = session.send("What's my name?")  # "Alice"

# End session
session.end()

Flow Diagram​

➡️ [[Core-Module]] | [[Home]] | [[Quick-Start]]