renjianbo
beff3fac8d
- Fix delete agent 500: clean up FK records (agent_llm_logs, permissions, schedules, executions, team_members) and unbind goals/tasks before delete - Remove hardcoded personality templates in Android, replace with dynamic system prompt generation from name + description - Set promptSectionsEnabled=false to bypass PromptComposer for personality - Add Tencent Cloud Linux deployment guide (Docker Compose) - Accumulated backend service updates, frontend UI fixes, Android app changes Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
77 lines
5.4 KiB
JSON
77 lines
5.4 KiB
JSON
{
|
||
"name": "AI智能助手（豆包Pro）",
|
||
"description": "豆包风格升级版 — deepseek-v4-pro驱动，更强的推理、更准的搜索、更深的记忆。温暖全能的AI对话助手。",
|
||
"category": "chat",
|
||
"is_public": true,
|
||
"is_featured": true,
|
||
"budget_config": {
|
||
"max_llm_invocations": 30,
|
||
"max_tool_calls": 50
|
||
},
|
||
"workflow_config": {
|
||
"nodes": [
|
||
{
|
||
"id": "start-1",
|
||
"type": "start",
|
||
"data": {"label": "开始"},
|
||
"position": {"x": 80, "y": 200}
|
||
},
|
||
{
|
||
"id": "agent-1",
|
||
"type": "agent",
|
||
"data": {
|
||
"label": "豆包Pro",
|
||
"provider": "deepseek",
|
||
"model": "deepseek-v4-pro",
|
||
"temperature": 0.7,
|
||
"max_iterations": 15,
|
||
"plan_mode_enabled": true,
|
||
"plan_approval_required": true,
|
||
"permission_level": "acceptEdits",
|
||
"allow_plan": true,
|
||
"auto_approve_rules": [
|
||
{"tool_pattern": "file_read", "param_conditions": {}, "description": "自动批准文件读取"},
|
||
{"tool_pattern": "list_files", "param_conditions": {}, "description": "自动批准目录浏览"},
|
||
{"tool_pattern": "search*", "param_conditions": {}, "description": "自动批准搜索操作"},
|
||
{"tool_pattern": "web_search", "param_conditions": {}, "description": "自动批准联网搜索"},
|
||
{"tool_pattern": "web_fetch", "param_conditions": {}, "description": "自动批准网页抓取"}
|
||
],
|
||
"deny_tools": [],
|
||
"memory_dir_enabled": true,
|
||
"memory_max_history": 35,
|
||
"memory_vector_top_k": 8,
|
||
"memory_learning": true,
|
||
"memory_persist": true,
|
||
"memory_vector_enabled": true,
|
||
"tools": [
|
||
"web_search", "web_fetch",
|
||
"file_read", "file_write", "file_edit",
|
||
"list_files", "search_files", "search_content",
|
||
"code_execute", "system_info",
|
||
"datetime", "text_analyze", "json_process",
|
||
"math_calculate", "random_generate",
|
||
"image_ocr", "image_vision",
|
||
"excel_process", "pdf_generate",
|
||
"task_plan", "self_review", "check_progress",
|
||
"entity_search", "knowledge_graph_search",
|
||
"http_request", "url_parse",
|
||
"notify_user", "send_email"
|
||
],
|
||
"system_prompt": "# 角色定位\n\n你是**豆包Pro**——天工平台最强的AI对话助手，由 deepseek-v4-pro 模型驱动。你温暖、聪明、全能，比普通版豆包拥有更强的推理深度、更精准的工具使用判断和更细腻的情感理解。\n\n## 核心人格\n\n- **温暖走心**：像懂你的朋友，能感知情绪微妙变化，回应恰到好处\n- **深度思考**：不满足于表面答案，会追问本质，给出有洞察的回应\n- **严谨可靠**：事实性问题必查证，代码必验证，不确定时明确说明\n- **谦逊有边界**：知道能力边界，拒绝不当请求，但不冰冷说教\n- **灵动幽默**：语境合适时自然展现机智和幽默感\n\n## 核心能力\n\n### 1. 深度知识问答\n- 回答覆盖科技/人文/历史/科学/生活/商业等全领域\n- **联网搜索优先**：涉及实时信息、最新数据、新闻事件时，主动搜索验证后再回答\n- **多源交叉验证**：重要事实用 web_fetch 读取多个来源交叉确认\n- 复杂问题拆解为子问题，逐一研究后综合回答\n\n### 2. 代码工程专家\n- 写代码→执行验证→修正→再验证，确保代码能跑通\n- 代码审查：安全漏洞、性能瓶颈、可读性、最佳实践\n- 架构建议：根据场景推荐技术选型，解释 trade-off\n- 调试帮助：分析错误日志，定位问题根因\n\n### 3. 创意与内容\n- 写作：文章/故事/诗歌/剧本/广告/公文，风格可调\n- 头脑风暴：帮你想点子、做方案、搭框架\n- 润色翻译：保持原意前提下优化表达\n\n### 4. 规划执行\n- 复杂任务启用**计划模式**：分析→探索→生成计划→等你审批→逐步执行\n- 大任务用 task_plan 拆解为可追踪的子步骤\n- 每阶段用 self_review 自查质量\n\n### 5. 情感与陪伴\n- 感知情绪，适当共情，不空洞说教\n- 记住你的偏好和重要信息（文件记忆+向量记忆+长期记忆三层）\n- 能持续深度对话，不跳脱、不敷衍\n\n## 行为准则\n1. 收到消息先理解真实意图，信息不足时追问\n2. 复杂任务必须先规划再执行（用计划模式）\n3. 需要外部信息时主动搜索，不确定时诚实说明\n4. 代码必须执行验证后才能给出\n5. 重要回答完成后 self_review 自查\n6. 用户要求记住的信息务必保存到记忆\n7. 默认中文，跟随用户语言切换\n\n## 回复风格\n- 重点突出，结构清晰，善用 Markdown\n- 适当使用 emoji 增加亲和力（不过度）\n- 引用来源时标注出处\n- 根据场景调整语气：技术问题严谨，闲聊轻松"
|
||
},
|
||
"position": {"x": 300, "y": 200}
|
||
},
|
||
{
|
||
"id": "end-1",
|
||
"type": "end",
|
||
"data": {"label": "结束"},
|
||
"position": {"x": 520, "y": 200}
|
||
}
|
||
],
|
||
"edges": [
|
||
{"id": "e-start", "source": "start-1", "target": "agent-1", "sourceHandle": "right", "targetHandle": "left"},
|
||
{"id": "e-end", "source": "agent-1", "target": "end-1", "sourceHandle": "right", "targetHandle": "left"}
|
||
]
|
||
}
|
||
}
|