From 6997edc82e1cd555187f982701d03ebc51e26bba Mon Sep 17 00:00:00 2001
From: shikihane <shikihane@foxmail.com>
Date: Sun, 1 Mar 2026 19:19:31 +0800
Subject: [PATCH] feat(agent): wire Media through agent pipeline (cherry-pick
 PR #555)

Add Media field to processOptions, pass msg.Media from inbound
messages through to BuildMessages and serializeMessages so
vision-capable LLMs receive image_url content parts.

Based on work by @as3k in sipeed/picoclaw#555.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---
 pkg/agent/context.go |  3 ++-
 pkg/agent/loop.go    | 14 ++++++++------
 2 files changed, 10 insertions(+), 7 deletions(-)

diff --git a/pkg/agent/context.go b/pkg/agent/context.go
index 6fccbaf53..8868d6bf4 100644
--- a/pkg/agent/context.go
+++ b/pkg/agent/context.go
@@ -465,10 +465,11 @@ func (cb *ContextBuilder) BuildMessages(
 	messages = append(messages, history...)
 
 	// Add current user message
-	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 00b0f096a..52a72d0f1 100644
--- a/pkg/agent/loop.go
+++ b/pkg/agent/loop.go
@@ -46,11 +46,12 @@ 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
+	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)
@@ -417,6 +418,7 @@ func (al *AgentLoop) processMessage(ctx context.Context, msg bus.InboundMessage)
 		Channel:         msg.Channel,
 		ChatID:          msg.ChatID,
 		UserMessage:     msg.Content,
+		Media:           msg.Media,
 		DefaultResponse: defaultResponse,
 		EnableSummary:   true,
 		SendResponse:    false,
@@ -509,7 +511,7 @@ func (al *AgentLoop) runAgentLoop(ctx context.Context, agent *AgentInstance, opt
 		history,
 		summary,
 		opts.UserMessage,
-		nil,
+		opts.Media,
 		opts.Channel,
 		opts.ChatID,
 	)