Losita/mai-bot
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

根据开发组建议重命名，移除多余的Prompt后缀

Browse Source
This commit is contained in:
UnCLAS-Prommer
2026-02-02 21:10:39 +08:00
parent b793a3d62b
commit 74b852dd8b
38 changed files with 27 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/bw_learner/expression_learner.py
View File

@@ -68,7 +68,7 @@ class ExpressionLearner:
# 学习用（开启行编号，便于溯源）
random_msg_str: str = await build_anonymous_messages(random_msg, show_ids=True)
prompt_template = prompt_manager.get_prompt("learn_style_prompt")
prompt_template = prompt_manager.get_prompt("learn_style")
prompt_template.add_context("bot_name", global_config.bot.nickname)
prompt_template.add_context("chat_str", random_msg_str)

2
src/bw_learner/expression_selector.py
View File

@@ -378,7 +378,7 @@ class ExpressionSelector:
reply_reason_block = ""
# 3. 构建prompt（只包含情境，不包含完整的表达方式）
prompt_template = prompt_manager.get_prompt("expression_evaluation_prompt")
prompt_template = prompt_manager.get_prompt("expression_evaluation")
prompt_template.add_context("bot_name", global_config.bot.nickname)
prompt_template.add_context("chat_observe_info", chat_context)
prompt_template.add_context("all_situations", all_situations_str)

2
src/bw_learner/jargon_explainer.py
View File

@@ -200,7 +200,7 @@ class JargonExplainer:
explanations_text = "\n".join(jargon_explanations)
# 使用LLM概括黑话解释
prompt_of_summarize = prompt_manager.get_prompt("jargon_explainer_summarize_prompt")
prompt_of_summarize = prompt_manager.get_prompt("jargon_explainer_summarize")
prompt_of_summarize.add_context("chat_context", lambda _: chat_context)
prompt_of_summarize.add_context("jargon_explanations", lambda _: explanations_text)
summarize_prompt = await prompt_manager.render_prompt(prompt_of_summarize)

6
src/bw_learner/jargon_miner.py
View File

@@ -193,7 +193,7 @@ class JargonMiner:
"- 请参考上一次推断的含义，结合新的上下文信息，给出更准确或更新的推断结果"
)
prompt1_template = prompt_manager.get_prompt("jargon_inference_with_context_prompt")
prompt1_template = prompt_manager.get_prompt("jargon_inference_with_context")
prompt1_template.add_context("bot_name", global_config.bot.nickname)
prompt1_template.add_context("content", str(content))
prompt1_template.add_context("raw_content_list", raw_content_text)
@@ -233,7 +233,7 @@ class JargonMiner:
return
# 步骤2: 仅基于content推断
prompt2_template = prompt_manager.get_prompt("jargon_inference_content_only_prompt")
prompt2_template = prompt_manager.get_prompt("jargon_inference_content_only")
prompt2_template.add_context("content", str(content))
prompt2 = await prompt_manager.render_prompt(prompt2_template)
@@ -275,7 +275,7 @@ class JargonMiner:
logger.debug(f"jargon {content} 推断1结果: {response1}")
# 步骤3: 比较两个推断结果
prompt3_template = prompt_manager.get_prompt("jargon_compare_inference_prompt")
prompt3_template = prompt_manager.get_prompt("jargon_compare_inference")
prompt3_template.add_context("inference1", json.dumps(inference1, ensure_ascii=False))
prompt3_template.add_context("inference2", json.dumps(inference2, ensure_ascii=False))
prompt3 = await prompt_manager.render_prompt(prompt3_template)

2
src/bw_learner/reflect_tracker.py
View File

@@ -72,7 +72,7 @@ class ReflectTracker:
# LLM Judge
try:
prompt_template = prompt_manager.get_prompt("reflect_judge_prompt")
prompt_template = prompt_manager.get_prompt("reflect_judge")
prompt_template.add_context("situation", str(self.expression.situation))
prompt_template.add_context("style", str(self.expression.style))
prompt_template.add_context("context_block", context_block)