Skip to content

docs: use Responses API in README MCP examples #436

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
mvanhorn wants to merge 1 commit into
base: main
Choose a base branch
from
Open

docs: use Responses API in README MCP examples #436

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
60 changes: 33 additions & 27 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -947,9 +947,11 @@ Review [Structured Output Documentation](https://platform.openai.com/docs/guides
## Tools
### Remote MCP (Model Context Protocol)

The Model Context Protocol (MCP) enables AI models to securely connect to external data sources and tools through standardized server connections. This OpenAI Swift library supports MCP integration, allowing you to extend model capabilities with remote tools and services.
The Model Context Protocol (MCP) enables AI models to securely connect to external data sources and tools through standardized server connections. This OpenAI Swift library supports MCP integration through the Responses API, allowing you to extend model capabilities with remote tools and services.

You can use the [MCP Swift library](https://github.com/modelcontextprotocol/swift-sdk) to connect to MCP servers and discover available tools, then integrate those tools with OpenAI's chat completions.
You can use the [MCP Swift library](https://github.com/modelcontextprotocol/swift-sdk) to connect to MCP servers and discover available tools, then integrate those tools with OpenAI's Responses API.

MCP tools are a Responses API feature and are not supported by Chat Completions (`ChatQuery`), whose `tools` parameter accepts function tools only.

#### MCP Tool Integration

Expand All @@ -970,13 +972,13 @@ let mcpTool = Tool.mcpTool(
)
)

let query = ChatQuery(
messages: [
.user(.init(content: .string("Search for Swift repositories on GitHub")))
],
let query = CreateModelResponseQuery(
input: .textInput("Search for Swift repositories on GitHub"),
model: .gpt4_o,
tools: [mcpTool]
)

let response = try await openAI.responses.createResponse(query: query)
```

**MCP Tool Properties**
Expand Down Expand Up @@ -1017,39 +1019,43 @@ let mcpTool = Tool.mcpTool(
)
)

// Use in chat completion
let query = ChatQuery(
messages: [.user(.init(content: .string("Help me search GitHub repositories")))],
// Use in the Responses API
let query = CreateModelResponseQuery(
input: .textInput("Help me search GitHub repositories"),
model: .gpt4_o,
tools: [mcpTool]
)

let chatResult = try await openAI.chats(query: query)
let response = try await openAI.responses.createResponse(query: query)
```

**MCP Tool Call Handling**

When using MCP tools, the model may generate tool calls that are executed on the remote MCP server. Handle MCP-specific output items in your response processing:

```swift
// Handle MCP tool calls in streaming responses
for try await result in openAI.chatsStream(query: query) {
for choice in result.choices {
if let outputItem = choice.delta.content {
switch outputItem {
case .mcpToolCall(let mcpCall):
print("MCP tool call: \(mcpCall.name)")
if let output = mcpCall.output {
print("Result: \(output)")
}
case .mcpApprovalRequest(let approvalRequest):
// Handle approval request if requireApproval is enabled
print("MCP tool requires approval: \(approvalRequest)")
default:
// Handle other output types
break
}
let response = try await openAI.responses.createResponse(query: query)

for outputItem in response.output {
switch outputItem {
case .mcpToolCall(let mcpCall):
print("MCP tool call: \(mcpCall.name)")
if let output = mcpCall.output {
print("Result: \(output)")
}
if let error = mcpCall.error {
print("Error: \(error)")
}
case .mcpListTools(let list):
let toolNames = list.tools.map(\.name).joined(separator: ", ")
print("MCP tools available from \(list.serverLabel): \(toolNames)")
case .mcpApprovalRequest(let approvalRequest):
// Handle approval request if requireApproval is enabled
print("MCP tool requires approval: \(approvalRequest.name)")
print("Arguments: \(approvalRequest.arguments)")
default:
// Handle other output types
break
}
}
```
Expand Down