From 4b589125ce8e2aa5351ec19c9bbaf0270c78c460 Mon Sep 17 00:00:00 2001
From: Zachary Guerrero <zack.grrr@gmail.com>
Date: Fri, 20 Feb 2026 14:33:02 -0800
Subject: [PATCH] feat: wire Media through agent pipeline to enable vision
 model support

- Add Media field to processOptions struct
- Pass msg.Media from InboundMessage to processOptions in processMessage
- Update BuildMessages to attach media to user message
- Pass opts.Media to BuildMessages instead of nil in runAgentLoop
- Enables vision-capable models (Gemini 2.0 Flash, etc.) to receive image_url content parts
---
 pkg/agent/context.go |  3 ++-
 pkg/agent/loop.go    | 20 +++++++++++---------
 2 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/pkg/agent/context.go b/pkg/agent/context.go
index 78f5f1ffa..3d3e4dff1 100644
--- a/pkg/agent/context.go
+++ b/pkg/agent/context.go
@@ -198,10 +198,11 @@ func (cb *ContextBuilder) BuildMessages(history []providers.Message, summary str
 
 	messages = append(messages, history...)
 
-	if strings.TrimSpace(currentMessage) != "" {
+	if strings.TrimSpace(currentMessage) != "" || len(media) > 0 {
 		messages = append(messages, providers.Message{
 			Role:    "user",
 			Content: currentMessage,
+			Media:   media,
 		})
 	}
 
diff --git a/pkg/agent/loop.go b/pkg/agent/loop.go
index add183aaf..66adafc2e 100644
--- a/pkg/agent/loop.go
+++ b/pkg/agent/loop.go
@@ -42,14 +42,15 @@ type AgentLoop struct {
 
 // processOptions configures how a message is processed
 type processOptions struct {
-	SessionKey      string // Session identifier for history/context
-	Channel         string // Target channel for tool execution
-	ChatID          string // Target chat ID for tool execution
-	UserMessage     string // User message content (may include prefix)
-	DefaultResponse string // Response when LLM returns empty
-	EnableSummary   bool   // Whether to trigger summarization
-	SendResponse    bool   // Whether to send response via bus
-	NoHistory       bool   // If true, don't load session history (for heartbeat)
+	SessionKey      string   // Session identifier for history/context
+	Channel         string   // Target channel for tool execution
+	ChatID          string   // Target chat ID for tool execution
+	UserMessage     string   // User message content (may include prefix)
+	Media           []string // Media URLs attached to the user message
+	DefaultResponse string   // Response when LLM returns empty
+	EnableSummary   bool     // Whether to trigger summarization
+	SendResponse    bool     // Whether to send response via bus
+	NoHistory       bool     // If true, don't load session history (for heartbeat)
 }
 
 func NewAgentLoop(cfg *config.Config, msgBus *bus.MessageBus, provider providers.LLMProvider) *AgentLoop {
@@ -313,6 +314,7 @@ func (al *AgentLoop) processMessage(ctx context.Context, msg bus.InboundMessage)
 		Channel:         msg.Channel,
 		ChatID:          msg.ChatID,
 		UserMessage:     msg.Content,
+		Media:           msg.Media,
 		DefaultResponse: "I've completed processing but have no response to give.",
 		EnableSummary:   true,
 		SendResponse:    false,
@@ -402,7 +404,7 @@ func (al *AgentLoop) runAgentLoop(ctx context.Context, agent *AgentInstance, opt
 		history,
 		summary,
 		opts.UserMessage,
-		nil,
+		opts.Media,
 		opts.Channel,
 		opts.ChatID,
 	)