Non-Streaming End-to-End MCP Tool Use with an AI Model

See more AI Examples

Demonstrates Ai.UseMcp in a complete, non-streaming Model Context Protocol (MCP) workflow. The example connects to an MCP server, registers the server's tools with an AI model via UseMcp, and asks a question the model can only answer well by calling one of those tools. When the model requests a tool, Chilkat automatically calls it on the MCP server, feeds the result back into the conversation, and continues — all inside a single (non-streaming) Ask.

Tip: Code to parse the returned JSON can be generated with Chilkat's online tool at https://tools.chilkat.io/jsonParse.

IncludeFile "CkMcp.pb"
IncludeFile "CkStringBuilder.pb"
IncludeFile "CkJsonObject.pb"
IncludeFile "CkAi.pb"

Procedure ChilkatExample()

    success.i = 0

    ;  Non-streaming, end-to-end MCP example.  It connects to an MCP server, hands the server's tools to
    ;  an AI model via Ai.UseMcp, and asks a question that the model can only answer by calling one of
    ;  those tools.  When the model requests a tool, Chilkat automatically calls it on the MCP server,
    ;  feeds the result back into the conversation, and continues -- all inside a single (non-streaming)
    ;  Ask.
    ;  
    ;  The MCP server used here is DeepWiki: a free, public, no-authentication Streamable HTTP server
    ;  that answers questions about public GitHub repositories.  Swap in any MCP server URL; if it
    ;  requires a bearer token, set mcp.AuthToken before Connect.

    ;  1) Connect to the MCP server.
    mcp.i = CkMcp::ckCreate()
    If mcp.i = 0
        Debug "Failed to create object."
        ProcedureReturn
    EndIf

    ;  mcp.AuthToken = "...";   // <-- only if the server requires a bearer token
    success = CkMcp::ckConnect(mcp,"https://mcp.deepwiki.com/mcp")
    If success = 0
        Debug CkMcp::ckLastErrorText(mcp)
        CkMcp::ckDispose(mcp)
        ProcedureReturn
    EndIf

    Debug "Connected to MCP server: " + CkMcp::ckServerName(mcp) + " (version " + CkMcp::ckServerVersion(mcp) + ")"

    ;  (Optional) Show the tools the server advertises.  This is the same set of tools Chilkat registers
    ;  with the model in the UseMcp call below.
    toolsJson.i = CkJsonObject::ckCreate()
    If toolsJson.i = 0
        Debug "Failed to create object."
        ProcedureReturn
    EndIf

    CkJsonObject::setCkEmitCompact(toolsJson, 0)
    success = CkMcp::ckListTools(mcp,toolsJson)
    If success = 1
        Debug "Available MCP tools:"
        Debug CkJsonObject::ckEmit(toolsJson)
        Debug "----"
    EndIf

    ;  2) Set up the AI conversation and register the MCP tools.
    ai.i = CkAi::ckCreate()
    If ai.i = 0
        Debug "Failed to create object."
        ProcedureReturn
    EndIf

    CkAi::setCkProvider(ai, "anthropic")
    ;  The API key should come from a secure source rather than being hard-coded.
    CkAi::setCkApiKey(ai, "AI_PROVIDER_API_KEY")
    CkAi::setCkModel(ai, "claude-sonnet-5")
    CkAi::setCkVerboseLogging(ai, 1)

    ;  Automatic tool use (MCP or JavaScript) requires a conversation -- it cannot be used with a
    ;  stateless query, so create one with NewConvo.
    success = CkAi::ckNewConvo(ai,"test_conversation","You are a helpful assistant.  Use the available tools when they can help answer the question.","")
    If success = 0
        Debug CkAi::ckLastErrorText(ai)
        CkMcp::ckDispose(mcp)
        CkJsonObject::ckDispose(toolsJson)
        CkAi::ckDispose(ai)
        ProcedureReturn
    EndIf

    ;  Register the connected server's tools with the model.  Each tool name is namespaced with the
    ;  "deepwiki" prefix so that tools from different servers (and JavaScript tools) can never collide --
    ;  e.g. "deepwiki_ask_question".  The Mcp object must stay connected and in scope for as long as its
    ;  tools are in use during the conversation.
    success = CkAi::ckUseMcp(ai,mcp,"deepwiki")
    If success = 0
        Debug CkAi::ckLastErrorText(ai)
        CkMcp::ckDispose(mcp)
        CkJsonObject::ckDispose(toolsJson)
        CkAi::ckDispose(ai)
        ProcedureReturn
    EndIf

    ;  Multiple MCP servers can be used in the same conversation.  Connect each server and call
    ;  UseMcp once per server, giving each a distinct prefix so their tool names cannot collide.

    ;  A prompt the model can only answer well by calling one of the MCP tools.
    success = CkAi::ckInputAddText(ai,"Use the available tools to look up the GitHub repository " + Chr(34) + "modelcontextprotocol/modelcontextprotocol" + Chr(34) + ", then give me a brief, sourced summary of which transports the Model Context Protocol defines.")
    If success = 0
        Debug CkAi::ckLastErrorText(ai)
        CkMcp::ckDispose(mcp)
        CkJsonObject::ckDispose(toolsJson)
        CkAi::ckDispose(ai)
        ProcedureReturn
    EndIf

    ;  3) Non-streaming Ask.  Any MCP tool the model requests is executed automatically inside this call
    ;  (Chilkat -> Mcp.CallTool -> result back into the transcript -> continue), so Ask returns only once
    ;  the model produces its final answer.
    success = CkAi::ckAsk(ai,"text")
    If success = 0
        Debug CkAi::ckLastErrorText(ai)
        CkMcp::ckDispose(mcp)
        CkJsonObject::ckDispose(toolsJson)
        CkAi::ckDispose(ai)
        ProcedureReturn
    EndIf

    ;  The final assistant answer (after any automatic tool calls).
    sbResponse.i = CkStringBuilder::ckCreate()
    If sbResponse.i = 0
        Debug "Failed to create object."
        ProcedureReturn
    EndIf

    success = CkAi::ckGetOutputTextSb(ai,sbResponse)
    If success = 0
        Debug CkAi::ckLastErrorText(ai)
        CkMcp::ckDispose(mcp)
        CkJsonObject::ckDispose(toolsJson)
        CkAi::ckDispose(ai)
        CkStringBuilder::ckDispose(sbResponse)
        ProcedureReturn
    EndIf

    Debug "Assistant answer:"
    Debug CkStringBuilder::ckGetAsString(sbResponse)
    Debug "----"

    ;  (Optional) Dump the full transcript, which now includes the tool call and the tool result that
    ;  were added automatically.
    convoJson.i = CkJsonObject::ckCreate()
    If convoJson.i = 0
        Debug "Failed to create object."
        ProcedureReturn
    EndIf

    CkJsonObject::setCkEmitCompact(convoJson, 0)
    success = CkAi::ckExportConvo(ai,"test_conversation",convoJson)
    If success = 0
        Debug CkAi::ckLastErrorText(ai)
        CkMcp::ckDispose(mcp)
        CkJsonObject::ckDispose(toolsJson)
        CkAi::ckDispose(ai)
        CkStringBuilder::ckDispose(sbResponse)
        CkJsonObject::ckDispose(convoJson)
        ProcedureReturn
    EndIf

    Debug "Full Conversation:"
    Debug CkJsonObject::ckEmit(convoJson)

    ;  4) Close the MCP session (best effort; releases the server-side session).
    success = CkMcp::ckClose(mcp)
    If success = 0
        Debug CkMcp::ckLastErrorText(mcp)
        CkMcp::ckDispose(mcp)
        CkJsonObject::ckDispose(toolsJson)
        CkAi::ckDispose(ai)
        CkStringBuilder::ckDispose(sbResponse)
        CkJsonObject::ckDispose(convoJson)
        ProcedureReturn
    EndIf



    CkMcp::ckDispose(mcp)
    CkJsonObject::ckDispose(toolsJson)
    CkAi::ckDispose(ai)
    CkStringBuilder::ckDispose(sbResponse)
    CkJsonObject::ckDispose(convoJson)


    ProcedureReturn
EndProcedure