feat: add JWT auth token support to LLM inference activities

action/llm_inference.go

@@ -15,6 +15,7 @@ type LLMInferenceInput struct {
 	UserPrompt    string                 `json:"user_prompt"`    // User message
 	Temperature   float64                `json:"temperature,omitempty"`   // LLM temperature (0-1)
 	MaxTokens     int                    `json:"max_tokens,omitempty"`    // Max output tokens
+	AuthToken     string                 `json:"auth_token,omitempty"`    // JWT token for authenticated endpoints
 }
 
 // LLMInferenceOutput is output from LLMInferenceActivity
@@ -53,7 +54,7 @@ func LLMInferenceActivity(ctx context.Context, in LLMInferenceInput) (LLMInferen
 	}
 
 	// Call LLM
-	logger.logf("info", "Calling LLM API (model=%s, prompt_len=%d)", in.Model, len(in.UserPrompt))
+	logger.logf("info", "Calling LLM API (model=%s, prompt_len=%d, auth=%v)", in.Model, len(in.UserPrompt), in.AuthToken != "")
 
 	response, err := client.CreateMessage(ctx, llm.MessageInput{
 		Model: statemachine.ModelSpec{
@@ -66,6 +67,7 @@ func LLMInferenceActivity(ctx context.Context, in LLMInferenceInput) (LLMInferen
 				Content: in.UserPrompt,
 			},
 		},
+		AuthToken: in.AuthToken,
 	})
 
 	if err != nil {
@@ -88,6 +90,7 @@ type LLMBatchInferenceInput struct {
 	SystemPrompt string                   `json:"system_prompt"`
 	Prompts      []string                 `json:"prompts"`       // List of user prompts
 	Temperature  float64                  `json:"temperature,omitempty"`
+	AuthToken    string                   `json:"auth_token,omitempty"`    // JWT token for authenticated endpoints
 }
 
 // LLMBatchInferenceOutput is output from batch inference