Merge branch 'dev' of https://github.com/MoFox-Studio/MoFox_Bot into dev
This commit is contained in:
@@ -29,10 +29,11 @@ logger = get_logger("proactive_thinking_executor")
|
||||
|
||||
# 决策 Prompt
|
||||
decision_prompt_template = Prompt(
|
||||
"""你的人设是:
|
||||
"""{time_block}
|
||||
你的人设是:
|
||||
{bot_personality}
|
||||
|
||||
现在是 {current_time},你正在考虑是否要在与 "{stream_name}" 的对话中主动说些什么。
|
||||
你正在考虑是否要在与 "{stream_name}" 的对话中主动说些什么。
|
||||
|
||||
【你当前的心情】
|
||||
{current_mood}
|
||||
@@ -84,7 +85,8 @@ decision_prompt_template = Prompt(
|
||||
|
||||
# 冒泡回复 Prompt
|
||||
simple_bubble_reply_prompt_template = Prompt(
|
||||
"""你的人设是:
|
||||
"""{time_block}
|
||||
你的人设是:
|
||||
{bot_personality}
|
||||
|
||||
距离上次对话已经有一段时间了,你决定主动说些什么,轻松地开启新的互动。
|
||||
@@ -113,10 +115,11 @@ simple_bubble_reply_prompt_template = Prompt(
|
||||
|
||||
# 抛出话题回复 Prompt
|
||||
throw_topic_reply_prompt_template = Prompt(
|
||||
"""你的人设是:
|
||||
"""{time_block}
|
||||
你的人设是:
|
||||
{bot_personality}
|
||||
|
||||
现在是 {current_time},你决定在与 "{stream_name}" 的对话中主动发起一次互动。
|
||||
你决定在与 "{stream_name}" 的对话中主动发起一次互动。
|
||||
|
||||
【你当前的心情】
|
||||
{current_mood}
|
||||
@@ -202,9 +205,12 @@ class ProactiveThinkingPlanner:
|
||||
if recent_messages:
|
||||
recent_chat_history = await message_api.build_readable_messages_to_str(recent_messages)
|
||||
|
||||
# 3. 获取bot人设
|
||||
# 3. 获取bot人设和时间信息
|
||||
individuality = Individuality()
|
||||
bot_personality = await individuality.get_personality_block()
|
||||
|
||||
# 构建时间信息块
|
||||
time_block = f"当前时间是 {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}"
|
||||
|
||||
# 4. 获取当前心情
|
||||
current_mood = "感觉很平静" # 默认心情
|
||||
@@ -242,7 +248,7 @@ class ProactiveThinkingPlanner:
|
||||
"interest_score": stream_data.get("stream_interest_score", 0.5),
|
||||
"recent_chat_history": recent_chat_history or "暂无最近聊天记录",
|
||||
"bot_personality": bot_personality,
|
||||
"current_time": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
|
||||
"time_block": time_block,
|
||||
"current_mood": current_mood,
|
||||
"last_decision": last_decision,
|
||||
}
|
||||
@@ -315,8 +321,8 @@ class ProactiveThinkingPlanner:
|
||||
last_decision_text += f"\n- 话题: {last_topic}"
|
||||
|
||||
decision_prompt = decision_prompt_template.format(
|
||||
time_block=context["time_block"],
|
||||
bot_personality=context["bot_personality"],
|
||||
current_time=context["current_time"],
|
||||
stream_name=context["stream_name"],
|
||||
current_mood=context.get("current_mood", "感觉很平静"),
|
||||
stream_impression=context["stream_impression"],
|
||||
@@ -378,6 +384,7 @@ class ProactiveThinkingPlanner:
|
||||
|
||||
if action == "simple_bubble":
|
||||
reply_prompt = simple_bubble_reply_prompt_template.format(
|
||||
time_block=context["time_block"],
|
||||
bot_personality=context["bot_personality"],
|
||||
current_mood=context.get("current_mood", "感觉很平静"),
|
||||
stream_impression=context["stream_impression"],
|
||||
@@ -387,8 +394,8 @@ class ProactiveThinkingPlanner:
|
||||
)
|
||||
else: # throw_topic
|
||||
reply_prompt = throw_topic_reply_prompt_template.format(
|
||||
time_block=context["time_block"],
|
||||
bot_personality=context["bot_personality"],
|
||||
current_time=context["current_time"],
|
||||
stream_name=context["stream_name"],
|
||||
current_mood=context.get("current_mood", "感觉很平静"),
|
||||
stream_impression=context["stream_impression"],
|
||||
|
||||
Reference in New Issue
Block a user