0f9fd1b25c
chore(release): 公开发布适配——泄漏净化 + README 修订 + 发布脚本与排除清单
2026-08-22 12:25:59 +08:00
8c1dd407df
feat(workflow): 工作流库管理工具与编写技能,修复运行时 UI 多处问题
...
- 新增 list_workflows / save_workflow 工具:归档式创建与覆盖,
归档时完整结构校验、覆盖前备份可恢复、name 与目录名一致性检查
- 新增内置技能 workflow-authoring:WORKFLOW.md 编写指南 +
init/validate 脚本 + 节点详解与三个完整示例,skill_hints 挂载
- deactivate_workflow 挪工具循环特判层并广播摘牌快照,
修复模型自主退出后 slash 菜单与快捷卡片残留
- 工作流卡片消失改为退出动画:store exiting 状态保留快照播完动画
再清空,QuickDock hasContent 动画期间保持展开
- 运行期 workflow 通知 visibility 修正为 compact(原误走用户消息渲染)
- 激活/退出通知文案中性化,不向模型透露内部操作路径
- 修复空闲对话刷新后输入栏 token 圆环无数值(bootstrapRoute 补拉取)
- 修复 slash 激活工作流后输入栏 "/" 残留(跳转新对话前落盘草稿)
- 侧边栏工作流导航图标放大至 20px
2026-08-21 23:40:06 +08:00
2c2c5aba4d
fix(modules): 修复 Windows UTF-8 编码链(对照 Codex #7290 类问题)
...
- persistent_terminal 解码写死 GBK 改为 UTF-8 优先 + GBK 回退逐块自适应,GBK 边界感知 + _pending_bytes 暂存半字,进程结束冲刷
- start.py:Windows 默认 cmd.exe 启动后注入 chcp 65001,与 PYTHONIOENCODING=utf-8 对齐,消除输出/解码自相矛盾
- 5 处 subprocess text=True 补 encoding=utf-8(后台命令x3、MCP stdio、版本管理、病毒扫描、容器管理),照 server/status/git.py 示范
- main.py:补 PYTHONUTF8/PYTHONIOENCODING 兜底 + stdout/stderr reconfigure;修复 print 中被双重编码固化的 emoji
- docx 验证器 XML 改二进制读取,由 lxml 按声明自解码
2026-07-31 22:39:51 +08:00
3bb47ff857
docs(repo): 恢复追踪 AGENTS.md 与 agentskills/,AGENTS.md 完成脱敏
2026-07-30 22:43:01 +08:00
bb420b0d04
chore(repo): 公开发布准备——取消追踪非源码文件并重写配置文档
...
- 取消追踪 552 个文件(运行态数据、密钥历史、实验/归档目录、
agentskills/easyagent/demo/android-webview-app 等不公开内容)
- .gitignore 追加公开发布排除项与 cache/ 忽略
- 重写 README.md:mac 为主/开发中声明、web/host 双模式、Docker 镜像、
数据存储、settings.json/custom_models.json 配置说明
- .env.example 删除失效项(AGENT_API_* 等),标注推荐 settings.json
- 新增 config/custom_models.json.example(Kimi-K3 蓝本 + DeepSeek 示例)
- 修复 utils/perf_log.py 硬编码本机路径(改走 config.paths.LOGS_DIR)
2026-07-25 00:44:15 +08:00
4fcd0ccc4d
feat(multi-agent): 子智能体系统升级
...
- prompt改造:子智能体注入AGENTS.md/执行环境/工作区信息/skills/项目记忆
- 工具增加:read_skill/recall_project_memory/todo_create/todo_update_task/save_webpage
- 上下文压缩:深度压缩机制,记录current_context_tokens,默认150k阈值可配置
- 模型升级:sub_agent_models.json支持thinkmode_status/extra_parameter,与主智能体对齐
- 角色管理三层结构:源码树预设multi_agent_roles/ + 运行态host/web预设 + web按用户隔离
- 启动同步:initialize_system调用sync_preset_roles同步预设到host和web运行态
- 模式判断:API用session.host_mode,工具用data_dir路径判断,不依赖IS_HOST_MODE
- 前端:个人空间新增子智能体管理页(角色CRUD/压缩阈值/模型选择),复用个人空间样式
- 入口改造:登录页移除多智能体按钮,QuickMenu加模式切换项,运行中对话禁止切换
- 工具调整:多智能体模式create_sub_agent去掉timeout/deliverables_dir参数
- skill禁止:sub-agent-guide在多智能体模式下禁止阅读
- .agents/统一为.astrion/路径修复
2026-07-14 02:28:45 +08:00
c08c4ab141
docs(agentskills): 更新子智能体使用指南
...
- 强调阻塞/后台模式的唯一判断标准是运行期间有没有别的事情要做
- 强调大范围搜索/调查前必须先自己进行浅层搜索
- 重写示例:国产开源智能体调研、程序 bug 定位
- 拆分后台模式场景:单个子智能体用阻塞,多个子智能体用后台并等待通知
2026-07-11 20:40:27 +08:00
765196ea3c
feat(skills): host 模式 skill 归档目录迁移到运行态 agentskills
2026-07-09 18:42:42 +08:00
78544cb205
refactor: remove run_python tool, consolidate all execution to run_command
2026-06-07 16:23:01 +08:00
3bc6f718eb
feat(ui): refine sidebar workspace and code blocks
2026-05-28 03:53:49 +08:00
398ee6e760
feat(memory): refactor proactive memory and conversation recall
...
Refactor conversation persistence into workspace-scoped storage with host workspace legacy migration, current workspace filtering, and active host workspace state protection.
Add proactive recent-conversation prompt injection with personal-space controls, configurable item count, empty-conversation filtering, and frozen per-conversation prompt metadata to keep request caches stable.
Introduce conversation_search and conversation_review tools for current-workspace history: multi-keyword search, list mode, current conversation exclusion, message/tool counts, formatted model output, read/save review modes, long-review fallback, and .agents/review storage.
Move internal workspace artifacts under .agents, including uploads, skills, compact results, and reviews; update prompt file-tree visibility, compression output, upload paths, and hidden versioning exclusions.
Add create_skill workflow and a Skills tool category, move read_skill into Skills, validate/归档 skill folders, support host global skills and Docker/API private skills, and sync private skills into .agents/skills and skill prompts.
Update frontend tool icons, statuses, and expanded renderers for conversation recall and skills workflows; add workspace conversation and skills-manager tests.
2026-05-27 03:09:42 +08:00
ae9157c161
fix(skills): update frontend design skill and read_skill output
2026-05-26 02:02:14 +08:00
67fdb715f5
refactor(messages): 统一运行期消息插入为 [系统通知|source] 前缀的 user 消息
...
- 新增 server/chat_flow_task_support.inject_runtime_user_message 作为唯一入口,
统一处理 add_conversation + messages.insert + sender(user_message) 三件事
- 子智能体 / 后台 run_command / runtime_guidance / mode notice 全部走该入口,
role 一律 user,inline 仅作 metadata 标记,不再复用 system role 表达语义
- inline poll 移出 tool 执行循环,改为整轮 tool 完成后批量注入,
避免在 assistant.tool_calls 与 tool 序列之间插入 user 导致 API 报错
- 删除 _insert_completion_notice_message / _record_sub_agent_message
以及 build_messages 中 system→user 的回转分支
- 移除已废弃的 wait_sub_agent 工具残留(后端注册、前端图标/动画、
prompts、SKILL 文档;保留 sleep 的 wait_sub_agent_ids 参数)
- 前端取消对 is_auto_generated/auto_message_type 的过滤,
让运行期注入的 user 消息在历史和实时推送中正常渲染
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-23 02:01:36 +08:00
b151f8ff71
feat: add host workspace manager, mcp-tool-config skill, and UI enhancements
2026-04-28 13:04:51 +08:00
44560829be
feat: add agents-md-writer skill scaffolding and references
2026-04-25 16:35:13 +08:00
4b7453ce4a
feat: 整合test/agents功能并更新配置
...
- 添加user_upload/和skills/到.gitignore
- 更新terminal-guide skill说明文档
- 添加logo图片
2026-04-13 17:47:19 +08:00
757718c492
feat: add user turn tool protection to shallow compression
...
- Add shallow_compress_keep_user_turn_tools config (default: 3)
- Add is_auto_generated metadata to system messages (sub_agent/background_command)
- Fix sub_agent notification path to include auto_message marker
- Frontend: add input field in PersonalizationDrawer
- Backend: implement protection logic in _run_auto_shallow_compression
2026-04-12 18:48:08 +08:00
e742f4c1a5
docs: rebalance terminal vs run_command skill guidance
2026-04-08 16:40:17 +08:00
a04eca3aab
fix: unify background run_command notifications with sub-agent flow
2026-04-08 16:13:40 +08:00
cf6f056a93
docs: 终端skill补充完成后缀标记最佳实践与反例
2026-04-08 13:01:19 +08:00
ad8d3bb155
docs: 收紧子智能体skill为禁止代码编写并强化推荐场景
2026-04-08 13:01:12 +08:00
494edc53e9
docs(skill): document blocking and background sub-agent modes
2026-04-03 11:35:00 +08:00
5ab3acef9c
feat: 添加 terminal-guide 和 sub-agent-guide skills
...
- 新增 terminal-guide skill: 持久化终端使用指南
- 新增 sub-agent-guide skill: 子智能体使用指南
- 优化终端工具定义和执行逻辑
- 更新系统提示词以引用新 skills
- 添加 utils/__init__.py 模块初始化文件
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-03-16 21:17:02 +08:00
07be7a1061
feat: gracefully stop tool execution on user request
...
- Remove direct task.cancel() calls, use stop flag instead
- Monitor stop flag every 100ms during tool execution
- Cancel tool task immediately when stop flag is detected
- Return "命令执行被用户取消" as tool result with role=tool
- Save cancellation result to conversation history
- Prevent abrupt task termination, allow graceful shutdown
Changes:
- server/socket_handlers.py: Comment out pending_task.cancel()
- server/tasks.py: Comment out entry['task'].cancel()
- server/chat_flow_tool_loop.py: Add stop flag monitoring loop
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-03-08 03:50:34 +08:00
89eeb449b5
chore: sync workspace updates
2026-02-23 01:20:41 +08:00
7472028997
feat: add skills framework and controls
2026-02-07 00:20:35 +08:00