Skip to main content
Search Wikipedia, pull articles, and summarise them with a single Agent using dedicated wiki tools. 
from praisonaiagents import Agent
from praisonai_tools import wiki_search, wiki_summary, wiki_page

agent = Agent(
    name="WikiResearcher",
    instructions="Search and summarize Wikipedia content.",
    tools=[wiki_search, wiki_summary, wiki_page],
)

agent.start("What is the history of artificial intelligence?")
Wikipedia research agent with search, page retrieval, and summarization tools.

​
Quick Start

1

Simple Usage

Attach the wiki tools and ask a knowledge question.
from praisonaiagents import Agent
from praisonai_tools import wiki_search, wiki_summary, wiki_page

agent = Agent(
    name="WikiResearcher",
    instructions="Search and summarize Wikipedia content.",
    tools=[wiki_search, wiki_summary, wiki_page],
)

agent.start("What is the history of artificial intelligence?")
2

With Configuration

Return structured knowledge with a Pydantic schema.
from praisonaiagents import Agent, Task, AgentTeam
from praisonai_tools import wiki_search, wiki_summary, wiki_page
from pydantic import BaseModel

class WikiKnowledge(BaseModel):
    topic: str
    summary: str
    key_facts: list[str]
    related_topics: list[str]

agent = Agent(
    name="WikiResearcher",
    instructions="Extract structured knowledge from Wikipedia.",
    tools=[wiki_search, wiki_summary, wiki_page],
)

task = Task(
    description="What is the history of artificial intelligence?",
    expected_output="Structured knowledge summary",
    agent=agent,
    output_pydantic=WikiKnowledge,
)

AgentTeam(agents=[agent], tasks=[task]).start()

​
How It Works

​
Simple

Agents: 1 — Single agent with Wikipedia tools handles search and content extraction.

​
Workflow

  1. Receive knowledge query
  2. Search Wikipedia articles
  3. Summarize findings

​
Setup

pip install praisonaiagents praisonai wikipedia
export OPENAI_API_KEY="your-key"

​
Run — Python

from praisonaiagents import Agent
from praisonai_tools import wiki_search, wiki_summary, wiki_page

agent = Agent(
    name="WikiResearcher",
    instructions="Search and summarize Wikipedia content.",
    tools=[wiki_search, wiki_summary, wiki_page]
)

result = agent.start("What is the history of artificial intelligence?")
print(result)

​
Run — CLI

praisonai "Explain quantum computing from Wikipedia" --tools wikipedia

​
Run — agents.yaml

framework: praisonai
topic: Wikipedia Research
roles:
  wiki_researcher:
    role: Wikipedia Research Specialist
    goal: Extract and summarize Wikipedia content
    backstory: You are an expert at finding knowledge
    tools:
      - wiki_search
      - wiki_summary
      - wiki_page
    tasks:
      research:
        description: What is the history of artificial intelligence?
        expected_output: A comprehensive summary

praisonai agents.yaml

​
Serve API

from praisonaiagents import Agent
from praisonai_tools import wiki_search, wiki_summary, wiki_page

agent = Agent(
    name="WikiResearcher",
    instructions="You are a Wikipedia research agent.",
    tools=[wiki_search, wiki_summary, wiki_page]
)

agent.launch(port=8080)

curl -X POST http://localhost:8080/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "Tell me about the Roman Empire"}'

​
Advanced Workflow (All Features)

Agents: 1 — Single agent with memory, persistence, structured output, and session resumability.

​
Workflow

  1. Initialize session for knowledge tracking
  2. Configure SQLite persistence for research history
  3. Search and extract with structured output
  4. Store findings in memory for follow-up queries
  5. Resume session for continued research

​
Setup

pip install praisonaiagents praisonai wikipedia pydantic
export OPENAI_API_KEY="your-key"

​
Run — Python

from praisonaiagents import Agent, Task, AgentTeam, Session
from praisonai_tools import wiki_search, wiki_summary, wiki_page
from pydantic import BaseModel

class WikiKnowledge(BaseModel):
    topic: str
    summary: str
    key_facts: list[str]
    related_topics: list[str]

session = Session(session_id="wiki-001", user_id="user-1")

agent = Agent(
    name="WikiResearcher",
    instructions="Extract structured knowledge from Wikipedia.",
    tools=[wiki_search, wiki_summary, wiki_page],
    memory=True
)

task = Task(
    description="What is the history of artificial intelligence?",
    expected_output="Structured knowledge summary",
    agent=agent,
    output_pydantic=WikiKnowledge
)

agents = AgentTeam(
    agents=[agent],
    tasks=[task],
    memory=True
)

result = agents.start()
print(result)

​
Run — CLI

praisonai "Explain AI history" --tools wikipedia --memory --verbose

​
Run — agents.yaml

framework: praisonai
topic: Wikipedia Research
memory: true
memory_config:
  provider: sqlite
  db_path: wiki.db
roles:
  wiki_researcher:
    role: Wikipedia Research Specialist
    goal: Extract structured knowledge
    backstory: You are an expert at finding knowledge
    tools:
      - wiki_search
      - wiki_summary
      - wiki_page
    memory: true
    tasks:
      research:
        description: What is the history of artificial intelligence?
        expected_output: Structured knowledge summary
        output_json:
          topic: string
          summary: string
          key_facts: array
          related_topics: array

praisonai agents.yaml --verbose

​
Serve API

from praisonaiagents import Agent
from praisonai_tools import wiki_search, wiki_summary, wiki_page

agent = Agent(
    name="WikiResearcher",
    instructions="Extract structured knowledge from Wikipedia.",
    tools=[wiki_search, wiki_summary, wiki_page],
    memory=True
)

agent.launch(port=8080)

curl -X POST http://localhost:8080/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "Tell me about Rome", "session_id": "wiki-001"}'

​
Monitor / Verify

praisonai "test wikipedia" --tools wikipedia --verbose

​
Cleanup

rm -f wiki.db

​
Features Demonstrated

FeatureImplementation
WorkflowMulti-tool Wikipedia research
DB PersistenceSQLite via memory_config
Observability--verbose flag
Toolswiki_search, wiki_summary, wiki_page
ResumabilitySession with session_id
Structured OutputPydantic WikiKnowledge model

​
Best Practices

Pass wiki_search, wiki_summary, and wiki_page together. The agent needs search to find an article, page to read it, and summary to condense it.
Reach for this agent when accuracy on established facts matters. For breaking news or current events, use the Web Search Agent instead.
Add output_pydantic with key_facts and related_topics so downstream code gets typed fields instead of a paragraph.
Wikipedia is one source. Pair with the Research Agent when a question needs perspectives beyond a single encyclopaedia entry.

Search the live web for current events.

Synthesise multiple sources into a report.