Enum MessageContent 

#[non_exhaustive]
pub enum MessageContent {
    Text(String),
    Json(Value),
    ToolCall {
        id: String,
        name: String,
        arguments: Value,
    },
    ToolResult {
        tool_call_id: String,
        content: String,
        is_error: bool,
    },
}

Content of a message, supporting text, JSON, and tool interactions.

Most messages use MessageContent::Text, but tool calling workflows require MessageContent::ToolCall and MessageContent::ToolResult.

Examples

use multi_llm::MessageContent;

// Plain text (most common)
let text = MessageContent::Text("Hello, world!".to_string());

// Structured JSON content
let json = MessageContent::Json(serde_json::json!({
    "intent": "greeting",
    "confidence": 0.95
}));

// Tool call from assistant
let tool_call = MessageContent::ToolCall {
    id: "call_123".to_string(),
    name: "get_weather".to_string(),
    arguments: serde_json::json!({"city": "London"}),
};

// Tool result to send back
let tool_result = MessageContent::ToolResult {
    tool_call_id: "call_123".to_string(),
    content: "Sunny, 22°C".to_string(),
    is_error: false,
};

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

Text(String)

Plain text content.

This is the most common content type, used for regular conversation.

Json(Value)

Structured JSON content.

Useful for passing structured data that doesn’t fit the tool calling model, or for internal processing of parsed LLM outputs.

ToolCall

Tool/function call request from the assistant.

When the LLM decides to call a tool, it returns this content type. Your application should execute the tool and return a MessageContent::ToolResult.

Fields

id: String

Unique identifier for this tool call (generated by the LLM).

name: String

Name of the tool to invoke (must match a defined Tool).

arguments: Value

Arguments to pass to the tool as JSON.

ToolResult

Result from executing a tool.

After executing a tool call, send the result back using this content type. The tool_call_id must match the id from the corresponding ToolCall.

Fields

tool_call_id: String

ID of the tool call this result responds to.

content: String

Result content from tool execution (typically a string or JSON string).

is_error: bool

Whether the tool execution failed. If true, content contains error details.

Trait Implementations

impl Clone for MessageContent

fn clone(&self) -> MessageContent

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for MessageContent

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for MessageContent

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Display for MessageContent

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl PartialEq for MessageContent

fn eq(&self, other: &MessageContent) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for MessageContent

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for MessageContent