skills
/
deerflow2
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
1,865 Commits 5 Branches 8 Tags 57 MiB
Commit Graph

1865 Commits

Author SHA1 Message Date
肖应宇 f06a3fb49e dev: 测试标识标题 2026-04-08 16:54:04 +08:00
肖应宇 8dc57763f5 fix(threads): 避免隐式订阅 events 流 2026-04-08 16:54:04 +08:00
肖应宇 f87f185bbd build: 前端构建通过 2026-04-08 13:51:14 +08:00
肖应宇 90e63422da refactor(frontend): rename SELECT_SKILL to SELECT_SKILLS 2026-04-08 13:29:43 +08:00
肖应宇 661c401b92 chore(scripts): add frontend standalone deploy script 2026-04-08 13:12:34 +08:00
肖应宇 d3a6669d0a feat(skills): update image/video generation skill assets and scripts 2026-04-08 13:11:16 +08:00
肖应宇 10cf4f0b00 feat(frontend): switch skill messaging to selectedSkills payload 2026-04-08 13:09:47 +08:00
肖应宇 a8cfe1c42e refactor(frontend): 将会话状态标识从 show_reuse_welcome 重命名为 is_chatting 2026-04-08 11:44:42 +08:00
肖应宇 6d66cdd3f5 fix: 调整artifact iframe img间隙 2026-04-08 10:44:08 +08:00
肖应宇 4e95838f1f refactor(frontend): 将复用欢迎态路由参数统一为 show_reuse_welcome 2026-04-08 10:44:08 +08:00
肖应宇 d91e1ea999 fix: 修复开发模式的docker启动选项，修改的docker路由 2026-04-08 09:58:10 +08:00
肖应宇 43471aacd1 feat: 删除不需要的TodoList组件 2026-04-07 16:37:02 +08:00
肖应宇 4e2ff8b5be feat: 精简showWelcomeStyle和isNewThread断言。实现在/[thread_id]?xclaw_used=true时，显示欢迎页 2026-04-07 16:35:37 +08:00
肖应宇 1606d79bcb fix: 收紧threadId的类型为string，删除无用isNewRoute变量，合并状态判断showInputBox至showWelcomeStyle 2026-04-07 16:08:42 +08:00
肖应宇 7012693802 feat(03): align workspace visual layer with legacy baseline 2026-04-07 14:34:22 +08:00
肖应宇 981bb8f005 feat(05): harden e2e suite with explainable skip strategy 2026-04-07 14:34:09 +08:00
肖应宇 643b61d15a feat(04): stabilize iframe messaging and markdown export flows 2026-04-07 14:34:01 +08:00
肖应宇 931c418c87 docs(milestone): archive v1.0 roadmap requirements and state 2026-04-07 14:32:25 +08:00
肖应宇 5a4da6e581 docs(05): add context plan summary and UAT artifacts 2026-04-07 14:32:05 +08:00
肖应宇 b45c0dba61 docs(04): add context plan summary and UAT artifacts 2026-04-07 14:31:44 +08:00
肖应宇 821ca6a46b docs(03): record gap-closure plan summary and updated UAT 2026-04-07 14:31:19 +08:00
肖应宇 d12fb11993 docs(01): sync research notes and remove stale validation draft 2026-04-07 14:30:53 +08:00
肖应宇 eb8f979e1b test(03): complete UAT - 1 passed, 3 issues 2026-04-07 13:48:12 +08:00
肖应宇 a21cd310ee fix(frontend): align skills bootstrap contract to b412 content_ids shape 2026-04-07 13:35:46 +08:00
肖应宇 75f62e7c15 feat(frontend): align skills bootstrap flow to titan contract 2026-04-07 13:29:13 +08:00
肖应宇 8b3914a999 docs(03): capture context plan and execution summary 2026-04-07 13:17:39 +08:00
肖应宇 8248456121 test(02): complete UAT - 4 passed, 0 issues 2026-04-07 13:05:39 +08:00
肖应宇 b63dc43d57 docs(01): finalize phase planning artifacts in zh-CN 2026-04-07 12:58:44 +08:00
肖应宇 5087c582cc docs(02-01): complete phase execution metadata 
- 新增 02-SUMMARY 并记录验证偏差

- 更新 ROADMAP 与 REQUIREMENTS 进度
 2026-04-07 12:55:16 +08:00
肖应宇 c01ac7b8de test(02-01): 增加线程与 skills 合同回归测试 
- 新增 node:test 覆盖线程路由与 bootstrap 合同归一

- 更新 e2e 路由辅助与用例，移除 isnew 依赖
 2026-04-07 12:54:01 +08:00
肖应宇 034e35c880 feat(02-01): 统一 skills bootstrap 合同到 content_ids 
- 新增 content_id 到 content_ids 最小兼容归一层

- 调用侧统一发送 content_ids，移除双主合同路径
 2026-04-07 12:53:54 +08:00
肖应宇 af21bbd163 feat(02-01): 移除 isnew 路由逻辑并收敛线程单路径 
- 新会话仅由 /workspace/chats/new 路由控制

- 删除 isnew 参数分支并同步 iframe 跳转逻辑
 2026-04-07 12:53:49 +08:00
肖应宇 b7a837b2bf docs(02): create phase plan 
- 落实 D-01/D-02/D-03 的执行任务

- 绑定 MERGE-02, LOGIC-03, LOGIC-04 验证路径
 2026-04-07 12:41:15 +08:00
肖应宇 6397cbd5e1 docs(02): capture phase context 2026-04-07 12:38:48 +08:00
肖应宇 c574c41b2c docs(01-01): 完成 phase 01 执行与验证归档 
- 新增 SUMMARY 与 VERIFICATION

- 更新 STATE 与 REQUIREMENTS 进度
 2026-04-07 12:32:56 +08:00
肖应宇 7499a6992a docs(01-01): 建立 titan overlap 决策矩阵 
- 输出 keep/replace/hybrid 决策表

- 绑定 L0/L1/L2 与 Phase 2/3 输入
 2026-04-07 12:30:43 +08:00
肖应宇 92905bbe2f docs(01-01): 产出冲突清单与分级口径 
- 新增 conflict-inventory.csv 机器可消费清单

- 新增 conflict-inventory.md 审计说明
 2026-04-07 12:30:32 +08:00
肖应宇 d4cffcded2 docs(01-01): 固化 merge 与 titan 审计证据链 
- 增加可复现 Git 命令链

- 记录 merge 与 Titan overlap 证据摘要
 2026-04-07 12:30:24 +08:00
肖应宇 b7ccdc0f79 docs(01): research phase domain 2026-04-07 12:05:50 +08:00
肖应宇 9015696f83 docs: initialize project planning artifacts 2026-04-07 11:54:45 +08:00
肖应宇 6411b3d7a0 docs(codebase): generate codebase map 2026-04-07 11:37:10 +08:00
Titan 6b900ccb60 feat(logging): enhance logging configuration to support environment variable override for log level 2026-04-07 10:04:17 +08:00
Titan 0f607441c8 feat(chats): implement dynamic styling based on xclaw_used parameter 2026-04-05 23:39:44 +08:00
Titan b412b5193b feat(skills): support batch skill bootstrap via content_ids; clear parent dir once before per-skill write 2026-04-05 18:37:55 +08:00
Titan 5de7a2ab46 fix: centralize UUID generation and validate query thread_id for skillBootstrap; ensure single execution and correct thread dir 2026-04-05 18:37:55 +08:00
Titan 5aa38ee108 feat(skills): include uploads directory in skill scanning 2026-04-05 18:20:03 +08:00
Titan 66bdc951f8 feat(skills): add remote YAML bootstrap and materialization APIs 2026-04-05 18:16:05 +08:00
Chris Z 117fa9b05d
fix(channels): normalize slack allowed user ids (#1802) 
* fix(channels): normalize slack allowed user ids

* style(channels): apply backend formatter

---------

Co-authored-by: haimingZZ <15558128926@qq.com>
Co-authored-by: suyua9 <1521777066@qq.com>
 2026-04-05 18:04:21 +08:00
肖 28474c47cb
fix: avoid command palette hydration mismatch on macOS (#1563) 
# Conflicts:
#	frontend/src/components/workspace/command-palette.tsx

Co-authored-by: luoxiao6645 <luoxiao6645@gmail.com>
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
 2026-04-05 16:35:33 +08:00
thefoolgy 8049785de6
fix(memory): case-insensitive fact deduplication and positive reinforcement detection (#1804) 
* fix(memory): case-insensitive fact deduplication and positive reinforcement detection

Two fixes to the memory system:

1. _fact_content_key() now lowercases content before comparison, preventing
   semantically duplicate facts like "User prefers Python" and "user prefers
   python" from being stored separately.

2. Adds detect_reinforcement() to MemoryMiddleware (closes #1719), mirroring
   detect_correction(). When users signal approval ("yes exactly", "perfect",
   "完全正确", etc.), the memory updater now receives reinforcement_detected=True
   and injects a hint prompting the LLM to record confirmed preferences and
   behaviors with high confidence.

   Changes across the full signal path:
   - memory_middleware.py: _REINFORCEMENT_PATTERNS + detect_reinforcement()
   - queue.py: reinforcement_detected field in ConversationContext and add()
   - updater.py: reinforcement_detected param in update_memory() and
     update_memory_from_conversation(); builds reinforcement_hint alongside
     the existing correction_hint

Tests: 11 new tests covering deduplication, hint injection, and signal
detection (Chinese + English patterns, window boundary, conflict with correction).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(memory): address Copilot review comments on reinforcement detection

- Tighten _REINFORCEMENT_PATTERNS: remove 很好, require punctuation/end-of-string boundaries on remaining patterns, split this-is-good into stricter variants
- Suppress reinforcement_detected when correction_detected is true to avoid mixed-signal noise
- Use casefold() instead of lower() for Unicode-aware fact deduplication
- Add missing test coverage for reinforcement_detected OR merge and forwarding in queue

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-05 16:23:00 +08:00