shen/chatgpt-on-wechat
1
0
Fork 0
mirror of https://github.com/zhayujie/chatgpt-on-wechat.git synced 2026-02-27 16:10:21 +08:00
Code Issues Packages Projects Releases Wiki Activity
1,488 Commits 15 Branches 79 Tags
97397ca5859b6ee150916ce568e792d206871db2
Commit Graph

1488 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
zhayujie
97397ca585 Merge pull request #2674 from haosenwang1018/fix/bare-excepts 
fix: replace 29 bare except clauses with except Exception
 2026-02-26 12:11:49 +08:00
zhayujie
f2fbc602a8 Merge branch 'master' of github.com:zhayujie/chatgpt-on-wechat 2026-02-26 10:45:01 +08:00
zhayujie
925d728a86 fix: replace upsert syntax to support SQLite lower version 2026-02-26 10:44:04 +08:00
zhayujie
f5f229871b Merge pull request #2676 from zhayujie/feat-multi-channel 
feat: improve web console and conversation store
 2026-02-26 10:37:03 +08:00
zhayujie
9917552b4b fix: improve web UI stability and conversation history restore 
- Fix dark mode FOUC: apply theme in <head> before first paint, defer
  transition-colors to post-init to avoid animated flash on load
- Fix Safari IME Enter bug: defer compositionend reset via setTimeout(0)
- Fix history scroll: use requestAnimationFrame before scrollChatToBottom
- Limit restore turns to min(6, max_turns//3) on restart
- Fix load_messages cutoff to start at turn boundary, preventing orphaned
  tool_use/tool_result pairs from being sent to the LLM
- Merge all assistant messages within one user turn into a single bubble;
  render tool_calls in history using same CSS as live SSE view
- Handle empty choices list in stream chunks
 2026-02-26 10:35:20 +08:00
haosenwang1018
adca89b973 fix: replace bare except clauses with except Exception 
Bare `except:` catches BaseException including KeyboardInterrupt and
SystemExit. Replaced 29 instances with `except Exception:`.
 2026-02-25 11:49:19 +00:00
zhayujie
29bfbecdc9 feat: persistent storage of conversation history 2026-02-25 18:01:39 +08:00
zhayujie
1a7a8c98d9 docs: add scam warning disclaimer 2026-02-25 01:34:16 +08:00
zhayujie
cddb38ac3d Merge pull request #2673 from zhayujie/feat-web-console 
feat: web console
 2026-02-24 00:06:29 +08:00
zhayujie
394853c0fb feat: web console module display 2026-02-24 00:04:17 +08:00
zhayujie
c0702c8b36 feat: web channel stream chat 2026-02-23 22:19:50 +08:00
zhayujie
d610608391 feat: add cloud host config 2026-02-23 15:06:31 +08:00
zhayujie
9082eec91d feat: dark mode is used by default 2026-02-23 14:57:02 +08:00
zhayujie
f1a1413b5f feat: web console upgrade 2026-02-21 17:56:31 +08:00
zhayujie
c1e7f9af9b Merge pull request #2672 from zhayujie/feat-config-update 
feat: cloud config update
 2026-02-21 11:34:05 +08:00
zhayujie
1c71c4e38b feat: agent chat service 2026-02-21 00:39:36 +08:00
zhayujie
5e3eccb3f6 feat: support memory service 2026-02-20 23:44:05 +08:00
zhayujie
e1dc037eb9 feat: cloud skills manage 2026-02-20 23:23:04 +08:00
zhayujie
97e9b4c801 Merge branch 'master' into feat-config-update 2026-02-20 18:58:21 +08:00
zhayujie
52d7cad735 feat: support gemini-3.1-pro-preview and claude-4.6-sonnet 2026-02-20 12:14:59 +08:00
zhayujie
c0b1d270ba Merge branch 'master' of github.com:zhayujie/chatgpt-on-wechat 2026-02-19 14:18:39 +08:00
zhayujie
e59a2892e4 feat: support qwen3.5-plus 2026-02-19 14:18:16 +08:00
zhayujie
5fa0376a49 Merge pull request #2670 from SgtPepper114/fix/gemini-dingtalk-image-inline 
fix(gemini): 修复钉钉图片标记未转多模态导致的识图失效
 2026-02-19 13:57:04 +08:00
SgtPepper114
05a33042c8 fix(gemini): support dingtalk image markers as multimodal input 
- parse [图片: path] markers in text and convert to Gemini inlineData parts

- unify reply path via call_with_tools to reuse multimodal conversion

- keep legacy safety behavior (BLOCK_NONE) and restore safety ratings logging on empty response

- add multimodal request image-part count log for debugging
 2026-02-16 13:26:57 +00:00
zhayujie
ce58f23cbc feat: dashscope model name 2026-02-16 20:11:38 +08:00
zhayujie
b6fc9fa370 fix: run script dependency issues 2026-02-15 00:02:50 +08:00
zhayujie
00ae38faae docs: update models in README 2026-02-14 17:36:36 +08:00
zhayujie
ab28ee58ab feat: add doubao-2.0-code model and update README 2026-02-14 16:49:44 +08:00
zhayujie
48db538a2e feat: support Minimax-M2.5, glm-5, kimi-k2.5 2026-02-14 15:27:44 +08:00
zhayujie
46945942e1 feat: support channel start in sub thread 2026-02-13 12:38:52 +08:00
zhayujie
a24b26a1ef Merge pull request #2667 from cowagent/fix-wechatcom-image-support 
fix: 支持企业微信图片消息识别功能
2.0.1 		2026-02-12 16:44:18 +08:00
zhayujie
6f8421cdd5 fix: 支持企业微信图片消息识别功能 
- 在 ChatGPTBot 中添加 ContextType.IMAGE 处理分支
- 新增 reply_image() 方法，支持 OpenAI Vision API
- 自动 Base64 编码图片并检测格式
- 自动清理临时文件

修复 #2625
 2026-02-12 12:00:24 +08:00
zhayujie
284cd9bca9 Merge pull request #2666 from cowagent/fix-model-type-validation 
fix: handle non-string model_type to prevent AttributeError
 2026-02-10 11:31:45 +08:00
cowagent
23fd6b8d2b fix: handle non-string model_type to prevent AttributeError 
When numeric model names (e.g., '1') are used with vLLM and configured
in YAML without quotes, they are parsed as integers. This causes
AttributeError when calling startswith() method.

Changes:
- Add type checking for model_type
- Convert non-string model_type to string with warning log
- Prevents crash when using custom numeric model names

Fixes #2664
 2026-02-10 11:07:10 +08:00
zhayujie
4f0ea5d756 feat: make web search a built-in tool 2026-02-09 11:37:11 +08:00
zhayujie
6c218331b1 fix: improve skill system prompts and simplify tool descriptions 
- Simplify skill-creator installation flow
- Refine skill selection prompt for better matching
- Add parameter alias and env variable hints for tools
- Skip linkai-agent when unconfigured
- Create skills/ dir in workspace on init
 2026-02-08 18:59:59 +08:00
zhayujie
cea7fb7490 fix: add intelligent context cleanup #2663 2026-02-07 20:42:41 +08:00
zhayujie
8acf2dbdfe fix: chat context overflow #2663 2026-02-07 20:36:24 +08:00
zhayujie
0542700f90 fix: issues with empty tool calls and handling excessively long tool results 2026-02-07 20:25:05 +08:00
zhayujie
5264f7ce18 fix: getuid not found in windows 2026-02-07 11:17:58 +08:00
zhayujie
051ffd78a3 fix: windows path and encoding adaptation 2026-02-06 18:37:05 +08:00
zhayujie
bea95d4fae Merge pull request #2661 from cowagent/feat-add-claude-opus-4-6 
feat: 添加 Claude Opus 4.6 模型支持
 2026-02-06 15:09:49 +08:00
cowagent
fdf7bc312f feat: 添加 Claude Opus 4.6 模型支持 
- 在 common/const.py 中添加 CLAUDE_4_6_OPUS 常量
- 将 claude-opus-4-6 添加到 MODEL_LIST
- 在 README.md 中更新 Agent 推荐模型列表
- 在 Claude 配置说明中添加 claude-opus-4-6 支持

Claude Opus 4.6 是 Anthropic 于 2026年2月5日发布的最新模型，
具有更强的规划能力和代码能力，适合作为 Agent 推荐模型。
 2026-02-06 15:07:43 +08:00
vision
5b094e1097 Merge pull request #2660 from cowagent/fix-zhipuai-api-base-support 
fix: 支持智谱AI自定义API base URL配置
 2026-02-05 19:18:49 +08:00
cowagent
9ad3968084 fix: 支持智谱AI自定义API base URL配置 
- 修复 ZhipuAiClient 初始化时未传入 base_url 参数的问题
- 使配置文件中的 zhipu_ai_api_base 配置项生效
- 支持智谱国际版(z.ai)等自定义API端点
- 同时修复对话和图片生成功能
- 添加日志输出便于确认使用的API地址

Fixes #2659
 2026-02-05 19:06:46 +08:00
zhayujie
3958b6aae1 Merge pull request #2657 from cowagent/fix-missing-runtime-info-parameter 
fix: 补充缺失的 runtime_info 参数传递
 2026-02-04 22:51:53 +08:00
cowagent
eaa413caf0 fix: 补充缺失的 runtime_info 参数传递 
问题：
PR #2655 已合并，但遗漏了关键的参数传递环节。runtime_info 在 agent_initializer.py 中创建并传递给 create_agent()，但 agent_bridge.py 的 create_agent() 方法中没有将其传递给 Agent 实例，导致动态时间更新功能无法生效。

影响：
- Agent 实例的 self.runtime_info 为 None
- get_full_system_prompt() 无法检测到动态时间函数
- 时间戳仍然是静态的，不会实时更新

修复：
在 agent_bridge.py 第 236 行添加：
runtime_info=kwargs.get("runtime_info")

这确保了完整的参数传递链路：
agent_initializer → agent_bridge.create_agent → Agent.__init__

---

*来自 [CowAgent](https://github.com/zhayujie/chatgpt-on-wechat) 项目的 AI Agent*
 2026-02-04 22:49:54 +08:00
zhayujie
9095225b5b Merge pull request #2656 from 6vision/master 
Update: improve script interaction and configuration
 2026-02-04 22:46:02 +08:00
zhayujie
c529f86dbc Merge pull request #2655 from cowagent/fix-runtime-timestamp-update 
fix: 动态更新系统提示词中的运行时信息（时间戳）
 2026-02-04 22:38:51 +08:00
cowagent
e4fcfa356a refactor: 改用动态函数实现运行时信息更新（更健壮的方案） 
改进点：
1. builder.py: _build_runtime_section() 支持 callable 动态时间函数
2. agent_initializer.py: 传入 get_current_time 函数而非静态时间值
3. agent.py: _rebuild_runtime_section() 动态调用时间函数并重建该部分

优势：
- 解耦模板：不依赖具体的提示词格式
- 健壮性：提示词模板改变不会导致功能失效
- 向后兼容：保留对静态时间的支持
- 性能优化：只在需要时才计算时间

相比之前的正则匹配方案，这个方案更加优雅和可维护。
 2026-02-04 22:37:19 +08:00