Run a Dappier MCP server and toolkit-enabled chatbot inside a temporary, cloud-hosted Meshagent Room for testing or development. When the Room closes (period of inactivity), the server and chatbot automatically shut down — no local process needed. Required Environment Variables:
  • DAPPIER_API_KEY — Your Dappier API Key (from Dappier)
Steps:
  • pip install "meshagent[all]"
  • Sign up and authenticate with Meshagent
  • Start the MCP server agent in a new test Room: 
    meshagent service test \
  --room=test \
  --role=agent \
  --image=meshagent/mcp-dappier:latest \
  --env MESHAGENT_PORT=8001 \
  --env DAPPIER_API_KEY=YOUR_API_KEY_HERE \
  --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
  --name=mcp-dappier-service-test
    • Starts a Meshagent Room with a test MCP server. Room will auto-close when no activity is detected.
  • Join the Room with a toolkit-enabled chatbot: 
    meshagent chatbot join \
  --room=test \
  --agent-name=mcp-dappier \
  --name=mcp-dappier \
  --toolkit=mcp-dappier
    • Enables a chatbot in the Room that can use the Dappier toolkit. (Multiple toolkits per agent supported.)
    • The output of this command will include a browser-accessible link to your Room.
  • Visit the Room link in your browser and send a message to interact with the MCP Server tools!

​
Project Level Deployment (Persistent for Production)

Deploy MCP server and chatbot services at the Meshagent project level. Services auto-join any new Room created in your Meshagent project, ensuring continuous and consistent access to Dappier’s features in every workspace (no need to start/stop per Room). Steps:
  • Deploy the MCP server as a persistent agent: 
    meshagent service create \
  --role=agent \
  --image=meshagent/mcp-dappier:latest \
  --env MESHAGENT_PORT=8001 \
  --env DAPPIER_API_KEY=YOUR_API_KEY_HERE \
  --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
  --name=mcp-dappier-service
  • Deploy the persistent chatbot agent: 
    meshagent service create \
  --image="meshagent/cli:latest" \
  --port="num=9001 path=/agent liveness=/ type=meshagent.callable participant_name=mcp-dappier-chatbot" \
  --env="MESHAGENT_PORT=9001" \
  --name="mcp-dappier-chatbot-service" \
  --command="meshagent chatbot service --agent-name=mcp-dappier-chatbot --toolkit=mcp-dappier"
Now, every time a Room is created in your Meshagent project, the Dappier MCP server tools and chatbot are instantly available.

​
Tools Available

​
Dappier MCP Server

Enable fast, free real-time web search and access premium data from trusted media brands—news, financial markets, sports, entertainment, weather, and more. Build powerful AI agents with Dappier.

​
Available Tools

Tool NameShort Description
dappier_ai_recommendationsFetch AI-powered recommendations from Dappier by processing the provided query with a selected data model that tailors results to specific interests.
dappier_real_time_searchRetrieve real-time search data from Dappier by processing an AI model that supports two key capabilities: - Real-Time Web Search: Access the latest news, stock market data, weather, travel information, deals, and more using model am_01j06ytn18ejftedz6dyhz2b15.

​
Tool: dappier_ai_recommendations

Fetch AI-powered recommendations from Dappier by processing the provided query with a selected data model that tailors results to specific interests.
  • Sports News (dm_01j0pb465keqmatq9k83dthx34): Get real-time news, updates, and personalized content from top sports sources.
  • Lifestyle News (dm_01j0q82s4bfjmsqkhs3ywm3x6y): Current lifestyle updates, analysis, and insights.
  • iHeartDogs AI (dm_01j1sz8t3qe6v9g8ad102kvmqn): Dog care and health info from iHeartDogs.
  • iHeartCats AI (dm_01j1sza0h7ekhaecys2p3y0vmj): Cat care expert info from iHeartCats.
  • GreenMonster (dm_01j5xy9w5sf49bm6b1prm80m27): Conscious and compassionate living.
  • WISH-TV AI (dm_01jagy9nqaeer9hxx8z1sk1jx6): Sports, breaking news, politics, multicultural news, and more.
Parameters:
NameTypeDescription
data_model_idstringThe data model ID to use for recommendations (see above for options).
querystringThe user’s question or search phrase.
num_articles_refinteger(Optional) Minimum number of articles to return from the reference domain.
refstring(Optional) Site domain to prioritize recommendations.
search_algorithmstring(Optional) Search algorithm for retrieving articles.
similarity_top_kinteger(Optional) Number of top semantic-similarity matches to retrieve.
Retrieve real-time search data from Dappier by processing an AI model that supports two key capabilities:
  • Real-Time Web Search: Access the latest news, stock market data, weather, travel, deals, and more. Use model am_01j06ytn18ejftedz6dyhz2b15.
  • Stock Market Data: Real-time news and prices from financial APIs like Polygon.io; only when a stock ticker symbol is present. Use model am_01j749h8pbf7ns8r1bq9s2evrh.
Parameters:
NameTypeDescription
ai_model_idstringThe AI model ID to use (am_01j06ytn18ejftedz6dyhz2b15 for general, am_01j749h8pbf7ns8r1bq9s2evrh for stocks).
querystringSearch query for real-time information.

​
Additional Resources

Get started with powerful, real-time AI tools in collaborative rooms — with just one command!