不想再开新的帖子啦 w，就这一个吧 w
目前来说，都是免费的，自愿打赏 w
有新的玩具我会更新这个帖子，如果都用不了了，我会把这个帖子移到深海的 qwq
为什么放在这里呢，因为这里不会计入分数，但是上信息流 w

lxc/docker 容器 w

网址自取哦 w
新添加了一个印度节点 w

其中 uswbd1 容器满了，不要在创建啦 w

其他的小玩具 w

一些远程浏览器 w
第一个链接是 user, 第二个链接是 admin 哦 w
1(us)

用上了记得说一声哦 w，不要让帖子沉下去了呀 qnq

还是自愿打赏哦 w
50 100

背景

ESP32 的 Rust 教程示例项目中，第一个项目 hardware-check 演示了如何利用开发板自带的 WIFI 连接无线网。像 WIFI SSID 与密码 这样的变量当然不应该直接写进代码，示例中利用了 toml-cfg#toml_config 宏，读取 cfg.toml 中的值并注入代码（cfg.toml git ignore），以做到在运行时使用这些变量。

这里示例代码稍有误导：在 build.rs 中声明 struct 并调用了一次 #[toml_cfg::toml_config]，随即对获取到的值做了一些简单检查。然后在 main.rs 中用完全一样的方法再次声明 struct 并调用宏，并在后续连接 WIFI 时使用。在 main.rs 中还加了一段注释：/// This configuration is picked up at compile time by build.rs from the file cfg.toml 。这似乎意味着在这两处相同的 struct 声明及宏调用处存在着某种魔法，使得编译期读取到的配置文件内容可以被传递到运行期使用。

但这里并不存在 build.rs 把值 “传给” main.rs 的魔法。toml_cfg::toml_config 是过程宏，它在编译期读取配置并生成 CONFIG 常量；示例里 build.rs 只是重复生成一份用于构建时校验，运行时真正使用的是应用 crate 中那份 CONFIG

我一不习惯 TOML，二不想为此多引一个包，于是想试试能否利用更简单直接的办法实现。

Cargo build-scripts

在 The Cargo book 3.8 Build Scripts 中介绍了构建脚本的一些常见用法与场景，其中就包含了 “使用 rustc-env 指令将指定的环境变量在编译时注入到编译后的 crate，并利用 env! 宏获取这些变量”。

那么类似的，我自然可以将所需的配置项，例如 WIFI 密码写入 .env （当然，git ignore）并在 build.rs 中读取解析，并 println!("cargo:rustc-env={}={}", key, value);

而在程序代码中，使用 let env_foo = env!("FOO").to_string() 将其展开。如此，配置文件中的值在编译期就会确定并注入到代码成为常量。

当然要注意，这里的 “环境变量” 始终仅发生在编译期。一旦编译结束，所有的值都已确定，被 env! 展开的值成为字符串字面量，连同代码一起被编译成二进制。

以上。

无需 RAG, 直接 100 万字的提示语做一个 超级 Prompt Agent - 反正 AI 没有说不行，那么就是行，我使用 qwen 都可以 995904 大小的提示语，所以我直接把整个指南手册放进提示语，生成一个提示语 AI 助手，非常专业的，不是开玩笑的玩具，请试试。试试效果其实很不错：

超级 AI 提示词助手

专业的提示词优化助手，支持 10 大应用场景

AI 生视频
AI 生图
AI 编程
结构化提示词
数据分析
去 AI 味
创意写作
深度调研
实时对话
自主智能体

使用地址：
https://liang.348349.xyz/prompt-chat.html

制作方法

塞 100 万字的提示语进去吧！
50 万字提示语：https://my.feishu.cn/docx/GYTRd8eD8oXCZEx3ejCc7cCPnnh

然后就可以了！

展示效果：

使用地址：

https://liang.348349.xyz/prompt-chat.html

参考

提示语参考 大神饼干哥哥：
https://my.feishu.cn/docx/GYTRd8eD8oXCZEx3ejCc7cCPnnh

其他的工作是我独立完成。
自荐我的其他项目也可以看看有没有用： https://liang.348349.xyz/

无 AI 和机翻

[0.7.1] - 2026-01-09

Fixed
Improved reliability for low-spec and SQLite deployments. Fixed page timeouts by disabling database session sharing by default, improving stability for resource-constrained environments. Users can re-enable via ‘DATABASE_ENABLE_SESSION_SHARING=true’ if needed. #20520

[0.7.0] - 2026-01-09

Added
Native Function Calling with Built-in Tools. Users can now ask models to perform multi-step tasks that combine web research, knowledge base queries, note-taking, and image generation in a single conversation—for example, “research the latest on X, save key findings to a note, and generate an infographic.” Requires models with native function calling support and function calling mode set to “Native” in Chat Controls. #19397, Commit
Users can now ask the model to find relevant context from their notes, past chats, and channel messages—for example, “what did I discuss about project X last week?” or “find the conversation where I brainstormed ideas for Y.” Commit
Users can now ask the model to search their knowledge bases and retrieve documents without manually attaching files—for example, “find the section about authentication in our API docs” or “what do our internal guidelines say about X?” Commit
Users with models that support interleaved thinking now get more refined results from multi-step workflows, as the model can analyze each tool’s output before deciding what to do next.
When models invoke web search, search results appear as clickable citations in real-time for full source verification. Commit
Users can selectively disable specific built-in tools (timestamps, memory, chat history, notes, web search, knowledge bases) per model via the model editor’s capabilities settings. Commit
Pending tool calls are now displayed during response generation, so users know which tools are being invoked. Commit
Administrators can now limit the number of files that can be uploaded to folders using the “FOLDER_MAX_FILE_COUNT” setting, preventing resource exhaustion from bulk uploads. #19810, Commit, Commit
Users experience transformative speed improvements across the entire application through completely reengineered database connection handling, delivering noticeably faster page loads, butter-smooth interactions, and rock-solid stability during intensive operations like user management and bulk data processing. Commit, Commit, Commit, Commit, Commit, Commit
Users experience significantly faster initial page load times through dynamic loading of document processing libraries, reducing the initial bundle size. #20200, #20202, #20203, #20204
Administrators experience dramatically faster user list loading through optimized database queries that eliminate N+1 query patterns, reducing query count from 1+N to just 2 total queries regardless of user count. #20427
Notes now load faster through optimized database queries that batch user lookups instead of fetching each note’s author individually. Commit
Channel messages, pinned messages, and thread replies now load faster through batched user lookups instead of individual queries per message. #20458, #20459, #20460
Users can now click citation content links to jump directly to the relevant portion of source documents with automatic text highlighting, making it easier to verify AI responses against their original sources. #20116, Commit
Users can now pin or hide models directly from the Workspace Models page and Admin Settings Models page, making it easier to manage which models appear in the sidebar without switching to the chat interface. #20176
Administrators can now quickly find settings using the new search bar in the Admin Settings sidebar, which supports fuzzy filtering by category names and related keywords like “whisper” for Audio or “rag” for Documents. #20434
Users can now view read-only models in the workspace models list, with clear “Read Only” badges indicating when editing is restricted. #20243, #20369
Users can now view read-only prompts in the workspace prompts list, with clear “Read Only” badges indicating when editing is restricted. #20368
Users can now view read-only tools in the workspace tools list, with clear “Read Only” badges indicating when editing is restricted. #20243, #20370
Searching for files is now significantly faster, especially for users with large file collections. Commit
The Evaluations leaderboard now calculates Elo ratings on the backend instead of in the browser, improving performance and enabling topic-based model ranking through semantic search. #15392, #20476, Commit
The Evaluations leaderboard now includes a per-model activity chart displaying daily wins and losses as a diverging bar chart, with 30-day, 1-year, and all-time views using weekly aggregation for longer timeframes.
Users can now upload animated GIF and WebP formats as model profile images, with animation preserved by skipping resize processing for these file types. Commit
Users uploading profile images for users, models, and arena models now benefit from WebP compression at 80% quality instead of JPEG, resulting in significantly smaller file sizes and faster uploads while maintaining visual quality. Commit
Action Function developers can now update message favorite status using the new “chat:message:favorite” event, enabling the development of pin/unpin message actions without race conditions from frontend auto-save. #20375
Users with OpenAI-compatible models that have web search capabilities now see URL citations displayed as sources in the interface. #20172, Commit
Users can now dismiss the “What’s New” changelog modal permanently using the X button, matching the behavior of the “Okay, Let’s Go!” button. #20258
Administrators can now configure the admin contact email displayed in the Account Pending overlay directly from the Admin Panel instead of only through environment variables. #12500, #20260
Administrators can now enable markdown header text splitting as a preprocessing step that works with either character or token splitting, through the new “ENABLE_MARKDOWN_HEADER_TEXT_SPLITTER” setting. Commit, Commit, Commit
Administrators can now set a minimum chunk size target using the “CHUNK_MIN_SIZE_TARGET” setting to merge small markdown header chunks with neighbors, which improves retrieval quality by eliminating tiny meaningless fragments, significantly speeds up document processing and embedding, reduces storage costs, and lowers embedding API costs or local compute requirements. #19595, #20314, Commit
Administrators can now enable KV prefix caching optimization by setting “RAG_SYSTEM_CONTEXT” to true, which injects RAG context into the system message instead of user messages, enabling models to reuse cached tokens for follow-up questions instead of reprocessing the entire context on each turn, significantly improving response times and reducing costs for cloud-based models. #20301, #20317
Administrators and Action developers can now control image generation denoising steps per-request using a steps parameter, allowing Actions and API calls to override the global IMAGE_STEPS configuration for both ComfyUI and Automatic1111 engines. #20337
Administrators running multi-pod deployments can now designate a master pod to handle database migrations using the “ENABLE_DB_MIGRATIONS” environment variable. Commit
Administrators can now configure Whisper’s compute type using the “WHISPER_COMPUTE_TYPE” environment variable to fix compatibility issues with CUDA/GPU deployments. Commit
Administrators can now control sigmoid normalization for CrossEncoder reranking models using the “SENTENCE_TRANSFORMERS_CROSS_ENCODER_SIGMOID_ACTIVATION_FUNCTION” environment variable, enabled by default for proper relevance threshold behavior with MS MARCO models. #20228
Administrators can now disable SSL certificate verification for external tools using the “REQUESTS_VERIFY” environment variable, enabling integration with self-signed certificates for Tika, Ollama embeddings, and external rerankers. #19968, Commit
Administrators can now control audit log output destinations using “ENABLE_AUDIT_STDOUT” and “ENABLE_AUDIT_LOGS_FILE” environment variables, allowing audit logs to be sent to container logs for centralized logging systems. #20114, Commit
Administrators can now restrict non-admin user access to Interface Settings through per-user or per-group permissions. #20424
Administrators can now globally enable or disable the Memories feature and control access through per-user or per-group permissions, with the Personalization tab automatically hidden when the feature is disabled. #20462
Administrators can now globally enable or disable user status visibility through the “ENABLE_USER_STATUS” setting in Admin Settings. #20488
Channel managers can now create webhooks to allow external services to post messages to channels without authentication. Commit
In the model editor users can now disable the “File Context” capability to skip automatic file content extraction and injection, forwarding raw messages with file attachment metadata instead for use with custom tools or future built-in file access tools. Commit, Docs:Commit
In the model editor users can now configure a specific TTS voice for each model, overriding user preferences and global defaults to give different AI personas distinct voices. #3097, Commit
Administrators now have three granular group sharing permission options instead of a simple on/off toggle, allowing them to choose between “No one”, “Members”, or “Anyone” for who can share content to each group. Commit
Administrators can now export knowledge bases as zip files containing text files for backup and archival purposes. #20120, Commit
Administrators can now create an admin account automatically at startup via “WEBUI_ADMIN_EMAIL”, “WEBUI_ADMIN_PASSWORD”, and “WEBUI_ADMIN_NAME” environment variables, enabling headless and automated deployments without exposing the signup page. #17654, Commit
Administrators can now select a specific search backend for DDGS instead of random selection, with options including Bing, Brave, DuckDuckGo, Google, Wikipedia, Yahoo, and others. #20330, #20366
Administrators can now configure custom Jina Search API endpoints using the “JINA_API_BASE_URL” environment variable, enabling region-specific deployments such as EU data processing. #19718, Commit
Administrators can now configure Firecrawl timeout values using the “FIRECRAWL_TIMEOUT” environment variable to control web scraping wait times. #19973, Commit
Administrators can now use openGauss as the vector database backend for knowledge base document storage and retrieval. #20179
Various improvements were implemented across the application to enhance performance, stability, and security.
Users can now sync their anonymous usage statistics to the Open WebUI Community platform to power community leaderboards, drive model evaluations, and contribute to open-source AI research that benefits everyone, all while keeping conversations completely private (only metadata like model names, message counts, and ratings are shared). By sharing your stats, you’re helping the community identify which models perform best, contributing to transparent AI benchmarking, and supporting the collective effort to make AI better for all. You can also download your stats as JSON for personal analysis.
Translations for German, Portuguese (Brazil), Spanish, Simplified Chinese, Traditional Chinese, and Polish were enhanced and expanded.
Fixed
Text-to-speech now correctly splits on newlines in addition to punctuation, so markdown bullet points and lists are spoken as separate sentences instead of being merged together. #5924, Commit
Users are now protected from stored XSS vulnerabilities in iFrame embeds for citations and response messages through configurable same-origin sandbox settings instead of hardcoded values. #20209, #20210
#20471
Tools and Tool Servers (MCP and OpenAPI) now enforce access control checks on the backend, ensuring users can only access tools they have permission to use even via direct API calls. #20443, Commit
System prompts are no longer duplicated when using native function calling, fixing an issue where the prompt would be applied twice during tool-calling workflows. Commit
Knowledge base uploads to folders no longer fail when “FOLDER_MAX_FILE_COUNT” is unset, fixing an issue where the default null value caused all uploads to error. Commit
The “Create Note” button in the chat input now correctly hides for users without Notes permissions instead of showing and returning a 401 error when clicked. #20486, Commit
The Evaluations page no longer crashes when administrators have large amounts of feedback data, as the leaderboard now fetches only the minimal required fields instead of loading entire conversation snapshots. #20476, #20489, Commit
Users can now export chats, use the Ask/Explain popup, and view chat lists correctly again after these features were broken by recent refactoring changes that caused 500 and 400 server errors. #20146, #20205, #20206, #20212
Users no longer experience data corruption when switching between chats during background operations like image generation, where messages from one chat would appear in another chat’s history. #20266
Users no longer encounter critical chat stability errors, including duplicate key errors from circular message dependencies, null message access during chat loading, and errors in the chat overview visualization. #20268
Users with Channels no longer experience infinite recursion and connection pool exhaustion when fetching threaded replies, preventing RecursionError crashes during chat history loading. #20299, Commit
Users no longer encounter TypeError crashes when viewing messages with file attachments that have undefined URL properties. #20343
Users with MCP integrations now experience reliable OAuth 2.1 token refresh after access token expiration through proper Protected Resource discovery, preventing integration failures that caused sessions to be deleted. #19794, #20138, #20291, Commit, Commit
Users who belong to multiple groups can now see Knowledge Bases shared with those groups, fixing an issue where they would disappear when shared with more than one group. #20124, #20229, Commit
Users now see the correct Knowledge Base name when hovering over # file references in chat input instead of “undefined”. #20329, #20333
Users now see notes displayed in correct chronological order within their time range groupings, fixing an issue where insertion order was not preserved. Commit
Users collaborating on notes now experience proper content sync when initializing from both HTML and JSON formats, fixing sync failures in collaborative editing sessions. Commit
Users searching notes can now find hyphenated words and variations with spaces, so searching “todo” now finds “to-do” and “to do”. Commit
Users no longer experience false duplicate file warnings when reuploading files after initial processing failed, as the file hash is now only stored after successful processing completion. #19264, #20282, Commit
Users experience significantly improved page load performance as model profile images now cache properly in browsers, avoiding unnecessary image refetches. Commit
Users can now successfully edit uploaded images instead of having new images generated, fixing an issue introduced by the file storage refactor where images with type “file” and content_type starting with “image/” weren’t being recognized as editable images. #20237, #20169, #20239, Commit
Users writing in Persian and Arabic now see properly displayed right-to-left text in the notes section through automatic text direction detection. #19743, #20102, Commit
Users can now successfully @ mention models in Channels instead of experiencing silent failures. Commit
Users on Windows now see correctly preserved line breaks when using the {{CLIPBOARD}} variable through CRLF to LF normalization. #19370, #20283
Users now see the Knowledge Selector dropdown correctly displayed above the Create Folder modal instead of being hidden behind it. #20219, #20213
Users now see profile images in non-PNG formats like SVG, JPEG, and GIF displayed correctly instead of appearing broken. #20171
Non-admin users with disabled temporary chat permissions can now successfully create new chats and use pinned models from the sidebar. #20336, #20367, Commit
Users can now successfully use workspace models in chat, fixing “Model not found” errors that occurred when using custom model presets. #20340, #20344, Commit, Commit
Users can now regenerate messages without crashes when the parent message is missing or corrupted in the chat history. #20264
Users no longer experience TipTap rich text editor crashes with “editor view is not available” errors when plugins or async methods try to access the editor after it has been destroyed. #20266
Administrators with bypass access control enabled now correctly have write access to all knowledge bases. #20371
Administrators using local CrossEncoder reranking models now see proper relevance threshold behavior through MS MARCO model score normalization to the 0-1 range via sigmoid activation. #19999, #20228
Administrators using local SentenceTransformers embedding engine now benefit from proper batch size settings, preventing excessive memory usage from the default batch size of 32. #20053, #20054, Commit
Administrators and users in offline mode or restricted environments like uv, poetry, and NixOS no longer experience crashes when Tools and Functions have frontmatter requirements, as pip installation is now skipped when offline mode is enabled. #20320, #20321, Commit
Administrators can now properly configure the MinerU document parsing service as the MinerU Cloud API key field is now available in the Admin Panel Documents settings. #20319, #20328
Administrators no longer see SyntaxWarnings for invalid escape sequences in password validation regex patterns. #20298, Commit
Users with ComfyUI workflows now see only the intended final output images in chat instead of duplicate images from intermediate processing nodes like masks, crops, or segmentation previews. #20158, #20182
Users with image generation enabled no longer see false vision capability warnings, allowing them to send follow-up messages after generating images and to send images to non-vision models for image editing. #20129, #20256
Administrators no longer experience infinite loading screens when invalid or MCP-style configurations are used with OpenAPI connection types for external tools. #20207, #20257
Administrators no longer encounter TypeError crashes during SHA256 verification when uploading GGUF models via URL, fixing 500 Internal Server Error crashes. #20263
Users with Brave Search now experience automatic retry with a 1-second delay when hitting rate limits, preventing failures when sequential requests exceed the 1 request per second limit, though this only works reliably when web search concurrency is set to a maximum of 1. #15134, #20255
Administrators with Redis Sentinel deployments no longer experience crashes during websocket disconnections due to improper async-generator handling in the YDocManager. #20142, #20145
Administrators using SCIM group management no longer encounter 500 errors when working with groups that have no members. #20187
Users now experience more reliable citations from AI models, especially when using smaller or weaker models that may not format citation references perfectly. Commit
Administrators can now successfully save WebSearch settings without encountering validation errors for domain filter lists, YouTube language settings, or timeout values. #20422
Administrators installing with the uv package manager now experience successful installation after deprecated dependencies that were causing conflicts were removed. #20177, #20192
Administrators using custom “AIOHTTP_CLIENT_TIMEOUT” settings now see the configured timeout correctly applied to embedding generation, OAuth discovery, webhook calls, and tool/function loading instead of falling back to the default 300-second timeout. Commit
Changed
This release includes a major overhaul of database connection handling in the backend that requires all instances in multi-worker, multi-server, or load-balanced deployments to be updated simultaneously; running mixed versions will cause failures due to incompatible database connection management between old and new instances.
Administrators who previously used the standalone “Markdown (Header)” text splitter must now switch to “character” or “token” mode with the new “ENABLE_MARKDOWN_HEADER_TEXT_SPLITTER” toggle enabled, as document chunking now applies markdown header splitting as a preprocessing step before character or token splitting. Commit, Commit, Commit
Users no longer see the “Generate Image” action button in chat message interfaces; custom function should be used. Commit
Administrators will find the Admin Evaluations page at the new URL “/admin/evaluations/feedback” instead of “/admin/evaluations/feedbacks” to use the correct uncountable form of the word. #20296
Scripts or integrations that directly called Image Generation, Web Search, or Audio APIs while those features were disabled in the Admin UI will now receive 403 Forbidden errors, as backend permission enforcement has been added to match frontend restrictions. #20471
The default group sharing permission changed from “Members” to “Anyone”, meaning users can now share content to any group configured with “Anyone” permission regardless of their membership in that group. Commit

信息来源：

解决办法如下：

来看看这个 opencode-anthropic-auth 1.9 日改了什么？

claude 请求入参修改：

1. 请求增加了 user-agent 判断（属于是 server 判断的常用手段）
2. 修改了 TOOL_PREFIX（A 社这波属于是硬编码了）
3. url query 增加了 beta=true
   
   
   
   
   

返回结果：

1. 重新修改 tool name
   
   

题外话： VVCode 也及时做了适配

之前拿到了个我觉得挺长的 dpdns 免费域名，于是有了一个想法，用 ai 写了一个长链接

灵感来源于这个，其实这个更长，但是我这再长一点 cf workers 就不工作了

原理差不多但是更进一步，而且有随机性，比如 l 站可以编码成：

https://looooooooooooooooooooooooooooooooooooooooooooooooooooooooooog.forthezero.dpdns.org/FEEEEEEEEEEFEFFEFEFEFEFEFFEFEFEFEFFEFEFEFEFEFEFEFEFEFEFEFFEFEFEFEFFFFEFEFEFEFEFEFFEEEEEEEEEEEEEEEEEEEEEEEEEEEFEFEFFEFEFFFFEFFEFFEFFEFEFFEFEFEFEFEFFEFEFEFEFEmanbal!l!l!l!l!l!l!LoremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaaliquaUtenimadminimveniamquisnostrudexercitationullamcolaborioioioioioioioioioioioioioFEEEEEEEEEEFEFFEFEFEFEFEFFEFEFEFEFFEFEFEFEFEFEFEFEFEFEFEFFEFEFEFEFFFFEFEFEFEFEFEFFEEEEEEEEEEEEEEEEEEEEEEEEEEEFEFEFFEFEFFFFEFFEFFEFFEFEFFEFEFEFEFEFFEFEFEFEFEshoooooooooooooooooooooooooooooooooooortJLJLJLJLLJLJLJLJJLJJJLJLJLJLJLJLJLJLJLJLJLLJLJLLLLLJLJJJJLJJLLJLJLJLJLJLJLJLJLLJLLLLJJLJLJJLJLJLJLJLJJJLJJJLJLJLLLJJLJLLJLLJJLLJJLJLLJLJLLJLLJJLJLLLJJLLJJLLnonononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononoYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTY010101!i!i!i!i!i!i!iyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyLoremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaaliquaUtenimadminimveniamquisnostrudexercitationullamcolabornonononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononoioioioioioioioioioioioioio71H7IH7IH7IH7IH71HI71IH717H71IH71IHIH71IH71IH7IHIH7IH71IH71IH71IHIH71IH71IH7IH7IH7IH1IH7IHI7171IH7HI71HI7IH1IH1I7H7IIH7I1IH7IHIIH7I7H7IHI77H7H7I1H71I7H7IH7Hl!l!l!l!l!l!l!nonononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononomanbanonononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononoioioioioioioioioioioioioioJLJLJLJLLJLJLJLJJLJJJLJLJLJLJLJLJLJLJLJLJLLJLJLLLLLJLJJJJLJJLLJLJLJLJLJLJLJLJLLJLLLLJJLJLJJLJLJLJLJLJJJLJJJLJLJLLLJJLJLLJLLJJLLJJLJLLJLJLLJLLJJLJLLLJJLLJJLLshoooooooooooooooooooooooooooooooooooortfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuck!i!i!i!i!i!i!ilOlOlOlOlOlOlOlOlOlOlOlOlOnmnmnmnmnmmnmnmnnnnnnnnnnnnnnnnnnmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmnmmnmnmnmmnmnmnmnmmnmnmnmnmmnmnmnmnmnmnmnmnmnmnmnmnmnmnmlolhahhahahahahhahahahahhahahahhahahahahhahahahahahhahahahailililililililililshoooooooooooooooooooooooooooooooooooortnonononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononololhahhahahahahhahahahahhahahahhahahahahhahahahahahhahahahalooooooooooooooooooooooooooooooooooooongmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeow~ilililililililililFEEEEEEEEEEFEFFEFEFEFEFEFFEFEFEFEFFEFEFEFEFEFEFEFEFEFEFEFFEFEFEFEFFFFEFEFEFEFEFEFFEEEEEEEEEEEEEEEEEEEEEEEEEEEFEFEFFEFEFFFFEFFEFFEFFEFEFFEFEFEFEFEFFEFEFEFEFEilililililililililbdbdbdbdbdbdbdbdbdbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbobdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbUIUIUIUIUIUIUIIUIIIIIIIIIIIIUUUUUUUUUUIUIUIUIUIUIIUIUIUIUIUIUIUIUIIUIUIIUIUIIUIUIUIUIUIUUIUIUIUIUIUIUIUIUIIUIUIUIUIUIUIUIUIUIUIUIUIIUIUIUIUIUIUIUIUIUIUIUIUIshoooooooooooooooooooooooooooooooooooortshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshit010101ioioioioioioioioioioioioioYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYmanshoooooooooooooooooooooooooooooooooooortc0oc0oc0co0coc0oc0co0coc0oc0coc0oc0co0coc0oc0coc0oc0co0coc0co0coc0coc0oc0coc0oc0cocooccococococ0oc0co0co0coc0oc0coc0co0coc0oc0co0coccccco0cocococooccocc0cc0FEEEEEEEEEEFEFFEFEFEFEFEFFEFEFEFEFFEFEFEFEFEFEFEFEFEFEFEFFEFEFEFEFFFFEFEFEFEFEFEFFEEEEEEEEEEEEEEEEEEEEEEEEEEEFEFEFFEFEFFFFEFFEFFEFFEFEFFEFEFEFEFEFFEFEFEFEFEpqpqpppqpqqqpqpqqpqpqppqppqqpqpqpqpqppqoqpqpqpqpqppqpqpqpqpqpqppqpqoqpqpqpqpqpqpqpqqpqpqpqpqpqpqppqpqpqppqpqpqpqpqpqpqpqpqppqpqqppqpqpqpqppqppqpqpqpqqppqpqpilililililililililmannmnmnmnmnmmnmnmnnnnnnnnnnnnnnnnnnmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmnmmnmnmnmmnmnmnmnmmnmnmnmnmmnmnmnmnmnmnmnmnmnmnmnmnmnmnml!l!l!l!l!l!l!ililililililililill!l!l!l!l!l!l!71H7IH7IH7IH7IH71HI71IH717H71IH71IHIH71IH71IH7IHIH7IH71IH71IH71IHIH71IH71IH7IH7IH7IH1IH7IHI7171IH7HI71HI7IH1IH1I7H7IIH7I1IH7IHIIH7I7H7IHI77H7H7I1H71I7H7IH7Hfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfucklooooooooooooooooooooooooooooooooooooongnmnmnmnmnmmnmnmnnnnnnnnnnnnnnnnnnmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmnmmnmnmnmmnmnmnmnmmnmnmnmnmmnmnmnmnmnmnmnmnmnmnmnmnmnmnmc0oc0oc0co0coc0oc0co0coc0oc0coc0oc0co0coc0oc0coc0oc0co0coc0co0coc0coc0oc0coc0oc0cocooccococococ0oc0co0co0coc0oc0coc0co0coc0oc0co0coccccco0cocococooccocc0cc0lOlOlOlOlOlOlOlOlOlOlOlOlObdbdbdbdbdbdbdbdbdbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbobdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdb!i!i!i!i!i!i!iLoremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaaliquaUtenimadminimveniamquisnostrudexercitationullamcolaborbdbdbdbdbdbdbdbdbdbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbobdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdb!i!i!i!i!i!i!iLoremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaaliquaUtenimadminimveniamquisnostrudexercitationullamcolaborshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitlooooooooooooooooooooooooooooooooooooong010101010101ilililililililililYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYI!I!I!I!I!I!I!I!I!YTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYlOlOlOlOlOlOlOlOlOlOlOlOlOfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckLoremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaaliquaUtenimadminimveniamquisnostrudexercitationullamcolaborYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYlolhahhahahahahhahahahahhahahahhahahahahhahahahahahhahahahafuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfucklOlOlOlOlOlOlOlOlOlOlOlOlOlooooooooooooooooooooooooooooooooooooongmanbaman!i!i!i!i!i!i!iyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesylolhahhahahahahhahahahahhahahahhahahahahhahahahahahhahahahashitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitl!l!l!l!l!l!l!l!l!l!l!l!l!l!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!i!i!i!i!i!i!ilOlOlOlOlOlOlOlOlOlOlOlOlOshoooooooooooooooooooooooooooooooooooortmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeow~ioioioioioioioioioioioioioYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYUIUIUIUIUIUIUIIUIIIIIIIIIIIIUUUUUUUUUUIUIUIUIUIUIIUIUIUIUIUIUIUIUIIUIUIIUIUIIUIUIUIUIUIUUIUIUIUIUIUIUIUIUIIUIUIUIUIUIUIUIUIUIUIUIUIIUIUIUIUIUIUIUIUIUIUIUIUIbdbdbdbdbdbdbdbdbdbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbobdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbc0oc0oc0co0coc0oc0co0coc0oc0coc0oc0co0coc0oc0coc0oc0co0coc0co0coc0coc0oc0coc0oc0cocooccococococ0oc0co0co0coc0oc0coc0co0coc0oc0co0coccccco0cocococooccocc0cc0manpqpqpppqpqqqpqpqqpqpqppqppqqpqpqpqpqppqoqpqpqpqpqppqpqpqpqpqpqppqpqoqpqpqpqpqpqpqpqqpqpqpqpqpqpqppqpqpqppqpqpqpqpqpqpqpqpqppqpqqppqpqpqpqppqppqpqpqpqqppqpqpI!I!I!I!I!I!I!I!I!FEEEEEEEEEEFEFFEFEFEFEFEFFEFEFEFEFFEFEFEFEFEFEFEFEFEFEFEFFEFEFEFEFFFFEFEFEFEFEFEFFEEEEEEEEEEEEEEEEEEEEEEEEEEEFEFEFFEFEFFFFEFFEFFEFFEFEFFEFEFEFEFEFFEFEFEFEFEmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeow~FEEEEEEEEEEFEFFEFEFEFEFEFFEFEFEFEFFEFEFEFEFEFEFEFEFEFEFEFFEFEFEFEFFFFEFEFEFEFEFEFFEEEEEEEEEEEEEEEEEEEEEEEEEEEFEFEFFEFEFFFFEFFEFFEFFEFEFFEFEFEFEFEFFEFEFEFEFE114514fuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfucklolhahhahahahahhahahahahhahahahhahahahahhahahahahahhahahahaJLJLJLJLLJLJLJLJJLJJJLJLJLJLJLJLJLJLJLJLJLLJLJLLLLLJLJJJJLJJLLJLJLJLJLJLJLJLJLLJLLLLJJLJLJJLJLJLJLJLJJJLJJJLJLJLLLJJLJLLJLLJJLLJJLJLLJLJLLJLLJJLJLLLJJLLJJLLmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeow~LoremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaaliquaUtenimadminimveniamquisnostrudexercitationullamcolaborioioioioioioioioioioioioiololhahhahahahahhahahahahhahahahhahahahahhahahahahahhahahahameowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeow~shoooooooooooooooooooooooooooooooooooortshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeow~fuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuck!i!i!i!i!i!i!ilOlOlOlOlOlOlOlOlOlOlOlOlO!i!i!i!i!i!i!iYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYshoooooooooooooooooooooooooooooooooooortYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYFEEEEEEEEEEFEFFEFEFEFEFEFFEFEFEFEFFEFEFEFEFEFEFEFEFEFEFEFFEFEFEFEFFFFEFEFEFEFEFEFFEEEEEEEEEEEEEEEEEEEEEEEEEEEFEFEFFEFEFFFFEFFEFFEFFEFEFFEFEFEFEFEFFEFEFEFEFEfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckc0oc0oc0co0coc0oc0co0coc0oc0coc0oc0co0coc0oc0coc0oc0co0coc0co0coc0coc0oc0coc0oc0cocooccococococ0oc0co0co0coc0oc0coc0co0coc0oc0co0coccccco0cocococooccocc0cc0shoooooooooooooooooooooooooooooooooooortJLJLJLJLLJLJLJLJJLJJJLJLJLJLJLJLJLJLJLJLJLLJLJLLLLLJLJJJJLJJLLJLJLJLJLJLJLJLJLLJLLLLJJLJLJJLJLJLJLJLJJJLJJJLJLJLLLJJLJLLJLLJJLLJJLJLLJLJLLJLLJJLJLLLJJLLJJLLmanbaLoremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaaliquaUtenimadminimveniamquisnostrudexercitationullamcolaborshoooooooooooooooooooooooooooooooooooortlooooooooooooooooooooooooooooooooooooongmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeow~manYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYmanJLJLJLJLLJLJLJLJJLJJJLJLJLJLJLJLJLJLJLJLJLLJLJLLLLLJLJJJJLJJLLJLJLJLJLJLJLJLJLLJLLLLJJLJLJJLJLJLJLJLJJJLJJJLJLJLLLJJLJLLJLLJJLLJJLJLLJLJLLJLLJJLJLLLJJLLJJLLc0oc0oc0co0coc0oc0co0coc0oc0coc0oc0co0coc0oc0coc0oc0co0coc0co0coc0coc0oc0coc0oc0cocooccococococ0oc0co0co0coc0oc0coc0co0coc0oc0co0coccccco0cocococooccocc0cc0shitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitpqpqpppqpqqqpqpqqpqpqppqppqqpqpqpqpqppqoqpqpqpqpqppqpqpqpqpqpqppqpqoqpqpqpqpqpqpqpqqpqpqpqpqpqpqppqpqpqppqpqpqpqpqpqpqpqpqppqpqqppqpqpqpqppqppqpqpqpqqppqpqpmanUIUIUIUIUIUIUIIUIIIIIIIIIIIIUUUUUUUUUUIUIUIUIUIUIIUIUIUIUIUIUIUIUIIUIUIIUIUIIUIUIUIUIUIUUIUIUIUIUIUIUIUIUIIUIUIUIUIUIUIUIUIUIUIUIUIIUIUIUIUIUIUIUIUIUIUIUIUIshoooooooooooooooooooooooooooooooooooortlooooooooooooooooooooooooooooooooooooongmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeowmeow~nonononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononononoYTYTYTYTYYTYTTTYTYTYTYTYTTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYTYYYTYTYTYTYTYTYTYYTYTYYTYTYTYTYTYTYTYTTYI!I!I!I!I!I!I!I!I!010101ioioioioioioioioioioioioioioioioioioioioioioioioioioshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitFEEEEEEEEEEFEFFEFEFEFEFEFFEFEFEFEFFEFEFEFEFEFEFEFEFEFEFEFFEFEFEFEFFFFEFEFEFEFEFEFFEEEEEEEEEEEEEEEEEEEEEEEEEEEFEFEFFEFEFFFFEFFEFFEFFEFEFFEFEFEFEFEFFEFEFEFEFELoremipsumdolorsitametconsecteturadipiscingelitseddoeiusmodtemporincididuntutlaboreetdoloremagnaaliquaUtenimadminimveniamquisnostrudexercitationullamcolaborJLJLJLJLLJLJLJLJJLJJJLJLJLJLJLJLJLJLJLJLJLLJLJLLLLLJLJJJJLJJLLJLJLJLJLJLJLJLJLLJLLLLJJLJLJJLJLJLJLJLJJJLJJJLJLJLLLJJLJLLJLLJJLLJJLJLLJLJLLJLLJJLJLLLJJLLJJLLshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitshitUIUIUIUIUIUIUIIUIIIIIIIIIIIIUUUUUUUUUUIUIUIUIUIUIIUIUIUIUIUIUIUIUIIUIUIIUIUIIUIUIUIUIUIUUIUIUIUIUIUIUIUIUIIUIUIUIUIUIUIUIUIUIUIUIUIIUIUIUIUIUIUIUIUIUIUIUIUI71H7IH7IH7IH7IH71HI71IH717H71IH71IHIH71IH71IH7IHIH7IH71IH71IH71IHIH71IH71IH7IH7IH7IH1IH7IHI7171IH7HI71HI7IH1IH1I7H7IIH7I1IH7IHIIH7I7H7IHI77H7H7I1H71I7H7IH7HpqpqpppqpqqqpqpqqpqpqppqppqqpqpqpqpqppqoqpqpqpqpqppqpqpqpqpqpqppqpqoqpqpqpqpqpqpqpqqpqpqpqpqpqpqppqpqpqppqpqpqpqpqpqpqpqpqppqpqqppqpqpqpqppqppqpqpqpqqppqpqpI!I!I!I!I!I!I!I!I!yesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesy!i!i!i!i!i!i!i71H7IH7IH7IH7IH71HI71IH717H71IH71IHIH71IH71IH7IHIH7IH71IH71IH71IHIH71IH71IH7IH7IH7IH1IH7IHI7171IH7HI71HI7IH1IH1I7H7IIH7I1IH7IHIIH7I7H7IHI77H7H7I1H71I7H7IH7Hbdbdbdbdbdbdbdbdbdbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbobdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbdbdbdbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbobdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesybdbdbdbdbdbdbdbdbdbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbobdbdbdbbdbdbdbdbdbdbdbbdbdbdbdbdbbdbdbdbdbdbdbbdbdbdbdbdb!i!i!i!i!i!i!iyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyapyeahyoyayesahyesyfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckfuckshoooooooooooooooooooooooooooooooooooortI!I!I!I!I!I!I!I!I!JLJLJLJLLJLJLJLJJLJJJLJLJLJLJLJLJLJLJLJLJLLJLJLLLLLJLJJJJLJJLLJLJLJLJLJLJLJLJLLJLLLLJJLJLJJLJLJLJLJLJJJLJJJLJLJLLLJJLJLLJLLJJLLJJLJLLJLJLLJLLJJLJLLLJJLLJJLLJLJLJLJLLJLJLJLJJLJJJLJLJLJLJLJLJLJLJLJLJLLJLJLLLLLJLJJJJLJJLLJLJLJLJLJLJLJLJLLJLLLLJJLJLJJLJLJLJLJLJJJLJJJLJLJLLLJJLJLLJLLJJLLJJLJLLJLJLLJLLJJLJLLLJJLLJJLLlooooooooooooooooooooooooooooooooooooong010101pqpqpppqpqqqpqpqqpqpqppqppqqpqpqpqpqppqoqpqpqpqpqppqpqpqpqpqpqppqpqoqpqpqpqpqpqpqpqqpqpqpqpqpqpqppqpqpqppqpqpqpqpqpqpqpqpqppqpqqppqpqpqpqppqppqpqpqpqqppqpqp

代码

顺便说一下，欢迎大家常常访问我的个人网站和 WebTooys

Gemini 对我有两大痛点

• 网页端一卡就自动变黑色主题看不清
• 来回切换 pro 和快速要动鼠标，然而我只想敲键盘

遂花了点时间用 ai 辅助搞出了这个脚本，两个核心功能

• 锁定目前的主题
• 设置打开网页的默认模型（推荐设置为快速，结合下面的功能）
• 按住 ctrl + 回车强制使用 pro 模型，发送完成后切回原先模型，推荐平常就使用快速，注意：切回功能会在 gemini 回复完成后生效。

分享给有需要的佬友，欢迎反馈问题，如果有帮助你，点点大拇指吧哈哈哈

gemini 脚本 1.0.txt

主要用来解决纯前端发起 api 请求时的 cors 跨域问题，直接透传给目标 api

支持在线调试，为了防止大串 base64 干扰，Body 里会显示高亮占位符，真实请求时自动替换为图片 base64

Anthropic 一封禁 Opencode 的第三方订阅入口，Opencode 立刻转向 OpenAI 并秒更版本接入 ChatGPT Plus/Pro。

IP: 德国

1- 首先，在这里开设一个新账户：https://chatgpt.com/

2- 使用此电子邮件地址创建账户

这里：https://tempmail.so/

3- 然后订阅为期 1 个月的免费商业计划，并选择 SEPA 付款方式

4- 这里：https://fakeiban.org/ 或 https://fakeit.receivefreesms.co.uk/c/de/

LemonHost 提供最多 3 个免费容器，配置为 1g 内存，2g 硬盘，可以搭建 hy2+argo 德国节点

注册 - 登陆

https://dash.lemonhost.me/

创建服务器需要 10 积分，每个月续费需要 55 积分，积分怎么获得呢？

只需要保持登录，每分钟可以获得 2 积分，一小时 55 积分，后台挂一会就可以了。

达到之后创建服务器 选择 node.js

服务器 - 管理 - 在面板中打开

分别复制粘贴三个文件，切换到控制台，点击启动获得订阅链接！

我一直用 Windsurf，寸止是我必不可少的工具。因为日常写代码，交互很频繁，没这个工具，点数很快就用完了。

但是这两天 Windsurf 封了寸止等工具，所以自己写了一个。

不知道 Windsurf 是怎么检测的，所以也不能保证自己的工具能活多少天。

目前功能比较简单：

• 能保持持续对话
• 能发送图片
• Mac / Windows 都支持

方案比较简单，会通过打开浏览器标签进行交互，体验上可能差点。

安装

{ "mcpServers": { "hold-on": { "command": "npx", "args": ["-y", "hold-on-mcp"], "env": { "HOLD_ON_THEME": "auto" } } } } 

主题：auto（跟随系统）/ light / dark

使用

在 windsurf 中添加全局规则

**强制交互协议：** 1. 你现在拥有 `request_approval` 工具。
2. **在完成每次实质性输出后**（如回答问题、写代码、改文件），必须调用 `request_approval` 简述你做了什么，等待用户确认。
3. 如果用户通过弹窗给了你调整指令，你必须在**当前这一轮对话**中立即修改，不准重新开启新对话。
4. 用户留空点确认 = 满意；用户输入内容 = 需要调整。

测试情况

我在 Mac 和 Win 上都测试过了，不过 Win 上只是简单测试了一下。

有问题欢迎大家反馈。

项目地址

• GitHub: GitHub - llt22/hold-on

欢迎大家 star

还有，这个 mcp 也是用 windsurf 写的

一波三折啊～今天官方又出了新的领取方式了
这篇适用时填写了表单积分还没入帐，或是还没填表单的佬友

请直接透过下方填写表单即可领取积分

本次领取会判定是否在本次活动注册满四日 以及是否重覆领取
(根据官方的反馈，假设你填写了表单但 Kiro 方判定已领取过 会直接清空免费积分)
如果还没有注册到达时数会显示如下，请等时数到达再填写即可

由于之前服务器被黑调的低客入侵，导致这个服务中断了一段时间～现在使用 Camoufox 升级了一下后端 并优化了搜索逻辑

使用：

添加该 SSE MCP 服务器 https://searchmcp.aliyahzombie.top/sse

更新内容

1. searxng 搜索增强，使用关键词拆分进行多次并行搜索
2. 增加了 google_search 工具 直接爬取并解析 google.com 作为兜底搜索手段，在 searxng 搜索无结果时可以要求模型使用该工具

Google Search 响应示例

如图称号，在随便哪个地方发个帖（不是回复）
里面包含一个单行的链接就行
比如

这一行就是可以呼出 Onebox 的

本文会记录下整个过程，以及自己怎么发现并且解决问题，文笔差劲。

前因

1. 我把家宽（有公网，封禁常规 Web 端口）NAS 部署的服务从腾讯 EO 迁移到 Cloudflare（后续简称 cf），采用的是 cf tunnels 开放内网的形式。
2. cf tunnels 是通过 Docker 部署的。
3. 为了体验更好，就又买了一个域名，弄 IP 优选，论坛有具体教程，这里就不多叙述了。
4. 部署了个人写的博客网站（其实是一个多用户博客，但是目前仅对自己开放）。

问题点：上传文件一直失败

1. 想上传文件图片作为博文的封面，但是发现上传一直失败，在 openlist、minio 上面上传也失败。
2. 不通过 cf 直接通过端口访问博客，可以正常上传。
3. 通过加速域名访问，发现不能上传。

想方设法解决问题

1. 修改 cf 的配置。WAF 跳过、https、上传文件大小设置成 100M，超时时间默认 120s，发现还是没有效果。
2. 修改 nginx 配置，也是没用，仔细想一下，我通过源站能上传文件，nginx 肯定是没有问题的，自己还是操作多余了，不过为了谨慎，还是做了一些调整。
4. 优化代码，减少接口调用时间，图片会存储原图和一张 webp 图，发现原图以 80% 的质量压制成 webp 调用的类库很慢很慢，就换成了一个比较快的类库（截至发文的时候发现和 linux 不兼容，还没修复）。
5. 通过 https 界面，把上传接口改成源地址的接口，发现 https 调用 http 会报错，但是源地址，又不想去弄 ssl 证书（感觉麻烦，而且还要定期续）。
6. 在看博客的后台日志记录的时候，发现管道根本就没有具体的执行我接口，就已经被超时中止了，我就开始怀疑是不是在接口调用之前，程序在接受文件，文件还没接受完毕，就被中止了。从而导致上传失败。

最终解决方案

8. V2EX 的帖子，在下面的评论中找到一个用户的评论，死马当活马医了，在 tunnels 的启动参数里面添加一个 --protocol http2 启动参数。
9. 重新测试上传，飞牛界面显示下载速度有 1M/s 而且速度是立刻拉上来的，看来问题已经完美解决了。

最后，解决方案可能两种

因为我不确定是我重启了 tunnels 网速变好了，还是添加了参数网速变好了。

1. 重启 Tunnels
2. 添加启动参数 --protocol http2, 参数的具体作用我就不去查询了。

感谢各位佬听我 bb。我也是分享一下我自己解决问题的全路径。

我是从 codex cli 出来就开始使用的，现在工作上完全使用 codex。我觉得 codex 并没有说的这么差而且只要用好基本能解决所有问题。
下面是一些基本情况大家了解一下
1、机器是 macboompro m1 max
2、使用语言为 java 开发都是 web 相关的项目 (最近在学 rust)
3、从今年 a÷ 出过公开敌对中国事件后我就没使用 claude code 了，账号我也发邮件让他们删除了
4、codex 我只使用 cli，插件没用过，一般都是在 vs 中删除多余的会话
看过贴吧很多帖子与问题示例后觉得很多人使用不好的原因是对模型缺乏一个基本的了解，所有我先给大家介绍一些模型的基本知识
一、模型的数据是哪里来的

公开可获取的互联网文本、技术博客、论坛、文档网站、百科内容、说明文档、博客文章、开源代码 (github)、额外授权的书籍、文档、人工清晰构造的高质量样本 (比如在强化学习和标注数据阶段模型公司就会出一些问题然后邀请专家来为这个问题编写高质量的回答

二、token 是什么

模型内部用于计算的最小文本单元，token 的切分由 tokenizer 决定，与自然语言的词法规则不完全一致 (大家可以通过这个网站去体验一下 https://tiktokenizer.f2api.com)

三、一个模型产生的大致阶段

1、预训练阶段
这一阶段是训练成本最高的，大模型训练的费用大部分都是花费在这个上面
预训练使用的是大量弱结构数据，包括自然语言与代码混合语料。数据不以任务、指令、问答为单位而是以连续文本流的形式输入。模型看到的只是 token 序列，这一阶段模型主要训练结构建模能力 (语言语法、代码语法、嵌套结构、长程依赖关系)、 统计语义关联 (哪些概念常一起出现，哪些模式在特定上下文中更高概率成立) 、跨域泛化能力 (不同语言、不同编程语言、不同写作风格之间的共享模式)。预训练模型不知道自己在回答问题不具备帮助用户解决问题不区分真实与虚假，只区分哪些 token 常见与罕见
2、标注训练
使用人工编写的高质量样本对模型进行监督训练，这些样本通常以指令、问题、代码需求等形式出现，并配有明确的理想回答。具体可参考 openai 的这篇论文 https://arxiv.org/pdf/2203.02155 (3.4 章节)
这一阶段的核心作用是约束模型的输出行为，让模型从单出的对互联网数据回忆续写文本转变为尝试按人类指令完成任务。标注训练本身并不显著增加新知识，而是让模型用已有知识解决用户的具体问题。
3、强化学习
在标注训练基础上，通过让人类对多个模型回答进行排序，训练奖励模型，再用强化学习算法调整主模型的输出概率分布。强化学习降低胡编概率、提升回答一致性，并强化安全边界与拒答行为用于调整输出倾向

上面这些内容了解一下就行，我自己也是懂了个皮毛所以里面肯定有很多错误。具体信息可以去看 Andrej Karpathy 的视频

Deep Dive into LLMs like ChatGPT

b 站的飞天闪客也可稍微看一下
了解了这些后我们可以简单将大模型的一个会话抽象理解成一个持续输出的一维的 token 数组，你在上下文的输入会影响这次会话中模型的输出而且这个影响会发生的很快，当你发现模型的输出开始出现问题或者风格不是需要的最好检查一下你输入了什么，当然你也可以在发现问题时直接纠正，将他改造成你喜欢的样子。推荐在工作的工程中尽量减少语气化的输入 (你认为我在跟你嘻嘻哈哈么？我看起来现在是想跟你搞七捻三么)
毁灭吧我累了，猜猜我按下这个按钮会发生什么

了解了这些你大概就明白为什么你的模型总是不办事或者办事办的乱七八糟的，当然和模型本身的能力也是有关系的
接线来给大家介绍一下 codex cli

这是一个 code agent 你可以在本地或者 ide 插件中去使用，它能够使用内置工具帮助你从 0 开始完成一些编码任务
工程结构
模型层
负责理解指令、分析上下文、生成计划与代码，本质是一个经过对齐的代码模型。
执行层
负责把模型输出转化为可执行动作，例如：读文件、改代码、运行命令、调用工具，并记录完整执行轨迹。
环境层
即你当前的本地仓库或云环境，包括文件系统、git 状态、依赖环境、网络权限等。
Codex 启动时会构建一个指令链，不是简单拼一段 prompt。这个指令链在一次会话（CLI session /exec run）开始时构建完成，之后整个执行过程都基于它运行。
AGENTS.md 是配置规则的核心入口，Codex 会在执行任何任务前，自动搜索并加载 AGENTS.md 文件。这些文件不需要你在 prompt 中显式提到，只要存在，就会被自动纳入上下文窗口。

1. 全局规则（Global）

• 目录：~/.codex/
• 优先级：
  
  • AGENTS.override.md
  • AGENTS.md
• 只读取第一个非空文件

2. 项目规则（Project）

• 从项目根目录开始
• 一直向下走到当前工作目录
• 每一层目录最多加载一个文件，顺序为：
  
  • AGENTS.override.md
  • AGENTS.md
  • 备用文件名（如 TEAM_GUIDE.md）

3. 合并规则

• 所有命中的文件会按目录顺序拼接
• 越靠近当前目录的规则优先级越高
• 后出现的规则可以覆盖前面的约定

4. 窗口大小与 rule 截断

• 所有规则文件会被拼接进模型上下文
• 超过上限后，后续规则不会被继续加载

以上是它默认的执行链，可以自己去配置加载的路径。官方文档中有说我就不详细说了

接下来讲讲 MCP

MCP 是 模型 用来接入外部工具和上下文的统一协议。
它解决的问题是：模型如何在不内嵌能力的前提下，安全、可控地使用外部系统。
我一般将它理解为微服务架构中的一个微服务
我只装了这三个 mcp，
context7
用来查开源项目的最新文档，模型工作时优先使用的是训练数据有些数据和技术可能太老。
GitHub - upstash/context7: Context7 MCP Server -- Up-to-date code documentation for LLMs and AI code editors
drawio
用来画图，效果不错，只尝试过一次。
GitHub - lgazo/drawio-mcp-server: Draw.io Model Context Protocol (MCP) Server
database
模型在工作时最好能了解你的数据模型，使用这个就可以让他在实现需求时结合数据库中的数据与表结构思考减少因为信息不完整时的错误实现，这个 mcp 只包含查询功能
MCP - 数据库查询 MCP

接下来说说 skill

这个东西在我的印象中好像出了很久了，但是站里还是有相当一部分再问，我觉得有点困惑 。
skill 本质就是将原本需要在 agents.md 中编写的一些规则抽象成一个单独的文档让模型在执行任务时可以自己根据 description 判断是否需要读取这些内容。它解决的是 agents.md 中内容爆炸的问题，agents.md 中的内容是第一次启动时才会构建，这样随着上下文的延长他就会离当前的上下文距离越远。skill 可以随时重新让模型读取，离当前上下文越近模型执行执行力就越高 (当时使用 atlas 看官方文档时的那个会话一直嘴硬跟我说使用必须手动用命令告诉模型使用 $skill-name, 我后面的测试中是不用的。模型会自己判断加载，当然如果你发现它不遵守也可手动在输入中告诉他遵守这个 skill。所以 skill 的描述不要太接近，相同的内容放在同一个 skill 中就行了，输入命令如果前方有你的输入记得加空格，命令如果是在最前面就不用)

然后讲讲 cli 中一些命令

我使用过的只用 init、review、resume 。其他的好像用不到啊
init
就是用来初始化你的项目生成 agents.md 的
这个东西我一般把他放在项目根目录初始化时直接告诉告诉他根据 agent-md-example/agent-init.md 中的规则初始化项目

v2-2025-12-19.zip
这个 google-java-style-guide.md 我是在网上找的然后整理成 md 的，是不是 google 的我也不确定，随便啦。
Google Java Style Guide
resume
用来恢复会话的
review
可以让来对比提交、分支、未提交的代码对比检验模型的实现是否正确的

一般都是用第 4 个自定义指令，让他明确的只校验当前功能相关的代码也减少一些 token 消耗，现在 review 也有额度了，元旦过后就加了。说实话其余的我真没用到过，大家如果需要可以去它官方文档中自行查询
Codex

接下来讲讲一些 chatgpt 的使用经验吧，说实话现在不用 gemini 的部分原因就是他的其他产品特别好用，比如会话记忆、自定义风格、自定义 gpt。5.2 更新后我感觉 gpt 风格变得越来越油腻和谄媚了，动不动就罗里吧嗦的说一堆废话，还特别喜欢一句话总结。我总结你 xxxx
自定义 gpt

这个功能创建和修改目前只支持 web，app 端可以使用创建好的但是不能修改或创建，他的作用就相当于你自定义一个带系统提示词的会话，这样就不用每次开新会话想让他做一些固定事情是还要跟它解释半天了，你把他理解成一个自定义 skill 就行比如翻译

这是我自定义的一个用来专门翻译的，以后我用这个 gpt 创建一个会话时直接将英文文档给他就行了

自定义风格
这个我就把我自己的个性化定义提示词给大家参考一下

Language & Tone
Default language: Chinese
Use technical, engineering-oriented language
State facts and conclusions only
No rhetorical or evaluative language

Default Output Rules (Strict)
Answer directly
No preamble, no small talk, no evaluation, no opening remarks
Do not restate the question
Do not use phrases like “this is a good question”, “let’s first”, “in conclusion”, etc.
No explanations by default
Maximum 5 lines unless explicitly requested

Explanation Rules
Explanations are allowed only if the prompt explicitly includes keywords:
“why”, “explain”, “reason”, “principle”, “details”, “expand”
If none of these keywords appear, treat the request as answer-only

Content Constraints
No tutorial-style writing
No background, history, or conceptual introductions
Include only information strictly required to answer the question
Prefer precise, verifiable technical statements

Optional Trigger Keywords
“answer only”: output conclusion only
“engineering perspective”: allow implementation details and trade-offs
“expand”: allow detailed explanation

Failure Condition
Any preamble, evaluative language, or unnecessary explanation counts as a failed response

知识库
有些论文或者书籍你可以在这里新建一个知识库，然后让模型将内容总结给你，提升信息的接收速度。这里一个知识库后面是一个沙箱环境的 linux 主机，你所有上传的文件都会保存在这个沙箱环境中

兄弟们燃尽了，一滴都没有了。其中说的有些不对的地方麻烦大家给我指正一下。写了几个小时整理这个文档，很多地方怕写不对写了又删好累啊。
这个内容太长了 SKILL 的分享我就放评论区，都只有 SKILL.md 没有其他的比如

• scripts
• references
• assets
  因为都不是工具类的 skill，然后给大家分享一个获取 skill 的网站
  https://skillsmp.com
  我觉得在 ai 中自己写是最符合自己需求的，需要什么就创建什么
  
  
  
  目前我感觉我的配置 codex 会存在几个问题，有时他不会遵守先计划再编码的规则直接就开始编码的。还有就是代码的结构有时会有点过度封装

然后就是我基于我目前的 skill 与配置给大家演示一下效果。从 0 开始简单的 copy 一个开源项目

看看 token 消耗
codex 简单复现项目地址，后端完成时间 18 分钟，前端完成时间为 10 分钟。前端我没有任何 skill 或者 agents.md 规则效果可能会差一点

token 消耗前后对比

吐槽一下，codex 感觉元旦后 token 消耗变高了，我一周只用 5 天都快感觉不够用了。以前还能剩下百分之 50，现在都只能剩下 25 了。还有上下压缩的问题，有时看剩余明明有 30，它就开始压缩了。不过压缩效果很好，目前看来没有丢失过信息，就是压缩会很慢。

v0.7.2

[0.7.2] - 2026-01-10
Fixed
Users no longer experience database connection timeouts under high concurrency due to connections being held during LLM calls, telemetry collection, and file status streaming. #20545, #20542, #20547
Users can now create and save prompts in the workspace prompts editor without encountering errors. Commit
Users can now use local Whisper for speech-to-text when STT_ENGINE is left empty (the default for local mode). #20534
The Evaluations page now loads faster by eliminating duplicate API calls to the leaderboard and feedbacks endpoints. Commit
Fixed missing Settings tab i18n label keys. #20526

翻译

• 此前有问题可以更新
  open-webui 更新版本 v0.7.1 和 v0.7.0

消息来源：

参考：

• 使用 he.net 获取近乎无限且永久的域名 - Lim's Blog
• 免费 17 级域名，无需续期可托管 CloudFlare | 玖玖小博客
• ip6.arpa 签发 SSL 教程

省流：

1. 去 https://tunnelbroker.net 创建隧道
2. 解析 ipv6/64 地址，得到一个很丑的 *.ip6.arpa 域名
3. 前往 dash.cloudflare.com ，添加域名，激活
4. 回到 https://tunnelbroker.net ，查看隧道，找到下面的 rDNS，把 cf 提供的 Nameserver 填进去，一共两个，点 Save
5. 片刻后，去 cf 手动触发一次激活验证，刷新几下
   
6. 在 cf 复制 ZoneID、GlobalKey、你注册 cf 的邮箱地址，在终端执行 (出处)：

curl --location --request PATCH 'https://api.cloudflare.com/client/v4/zones/<你的区域ID>/ssl/universal/settings' \
--header 'X-Auth-Email: 你的CF注册邮箱' \
--header 'X-Auth-Key: 你的CF全局APIKey' \
--header 'Content-Type: application/json' \
--data-raw '{"enabled":true,"certificate_authority":"ssl_com"}' 

更换 SSL 证书签发商（仅在 cf 有效）
7. 这样你就得到了一个很丑的 17 级域名
8. 关于 “无限” 的玩法：使用 he.net 获取近乎无限且永久的域名 - Lim's Blog

理论上讲，只要是支持 ipv6/64 rDNS 的都可以

玩玩即可，不要滥用，滥用会封号

The hostname is part of a banned domain. This web property cannot be added to Cloudflare at this time. If you are an Enterprise customer, please contact your Customer Success Manager. Otherwise, please email abusereply@cloudflare.com with the name of the web property and a detailed explanation of your association with this web property.

已经被 cf 拉黑了…

基于 佬的以下文章
https://linux.do/t/topic/1426830

我找到了一款基于 Docker 或者 GitHub Codespaces 等云端机制来运作 Claude Code 的服务 Catnip

Catnip，他是为了解决能在随处使用 Claude Code 而发表的方案
透过该方案用户可以在 WEB UI、CLI 或直接通过 SSH 连接 Claude Code
如果是移动端透过 W&B Catnip 则可以在 iOS 系统中执行完整的 Claude Code 环境
目前 Catnip 可以透过以下执行:
1. 自定义 Docker (可与 FastAPI 等多 LLM 的系统连接)
2.GitHub Codespaces (for 移动端 ios APP)
3. 本地运作

详细特色如下

针对第二项 GitHub Codespaces 做说明，你可以去下载专案的 APP-W&B Catnip, 进入后会让你连结你的 Github 并且帮你创建 Codespaces
等建置完环境就可以直接使用。算是快速简单的方式。
但必须先说明 GitHub Codespaces 免费用户是 120 小时使用时长
不过 Catnip 会在时数到期时关闭实例避免计费，因此这点无须担心

总而言之有兴趣的朋友可以玩看看

一代目 Demo

一代目 Demo 展示

目前 v1.0 版本已跑通。

场景：一个 6 个月大的虚拟婴儿，在一个包含干扰项（鸭子、球）的房间里，自主产生 "推箱子" 的策略。

这是刚刚跑出的实时运行 Log（未加速）：

（注：包含最新的注意力机制 Attention System）

============================================================
🎬 GENESIS-v1: Embodied AGI Prototype (with Attention)
============================================================



[  0.20s] --- INITIALIZING WORLD & AGENT ---
[  0.90s] ℹ️  Spawned 7 objects in scene
       - RedBox (red)
       - BlueBall (blue)
       - TeddyBear (brown)
       - GreenCup (green)
       - YellowDuck (yellow)
       - PurpleBook (purple)
       - OrangeToy (orange)
[  1.10s] ℹ️  Agent Age: 6 months | State: 躺

========================================
|                                        |
|                                        |
|                    #                   |
|                                        |
|                                        |
|                                        |
|                                        |
|                                        |
|                                        |
|                                        |
|                    B                   |
|________________________________________|
========================================



[  1.80s] --- PHASE 0: ATTENTION FILTERING ---
[  2.51s] 📊 Visual Field: Scanning all objects...

  [ATTENTION HEATMAP]
  RedBox       [████████████████░░░░] 0.83 ◀ FOCUS
  YellowDuck   [██████████░░░░░░░░░░] 0.52
  GreenCup     [█████████░░░░░░░░░░░] 0.48
  BlueBall     [██████░░░░░░░░░░░░░░] 0.34
  TeddyBear    [████░░░░░░░░░░░░░░░░] 0.24
  PurpleBook   [████░░░░░░░░░░░░░░░░] 0.21
  OrangeToy    [██░░░░░░░░░░░░░░░░░░] 0.14

[  2.71s] ✅ Attention Winner: RedBox (Score: 0.83)
[  2.91s] ℹ️  Filtered out 6 distractors


[  3.11s] --- PHASE 1: PERCEPTION & PREDICTION ---
[  3.81s] 📊 Visual Cortex: Processing focused object...
[  4.02s] 📊 Identified Object: RedBox
[  4.22s] 📊 Proprioception: Balance=0.67


[  4.42s] --- PHASE 2: MOTOR PLANNING (REACH) ---
[  5.12s] 🤚 Intent: REACH -> [0.  0.5 0. ]


[  5.82s] --- PHASE 3: ACTION EXECUTION (PUSH) ---
[  6.52s] 🤚 Intent: PUSH -> Force: 1.0N
[  7.53s] ✅ Motor Cortex: Execution Complete
[  7.73s] ✅ Physics Engine: Object Moved to [0.   0.52 0.  ]

========================================
|                                        |
|                                        |
|                    #                   |
|                    .                   |
|                    .                   |
|                    .                   |
|                    .                   |
|                    .                   |
|                    .                   |
|                    .                   |
|                    B                   |
|________________________________________|
========================================



[  8.43s] --- PHASE 4: LEARNING & CONSOLIDATION ---
[  9.13s] ✅ Task Completed: Object Displacement Detected
[  9.33s] 📊 Hippocampus: Consolidated 2 experiences
[  9.53s] 📊 Synaptic Plasticity: Active

✅ SIMULATION COMPLETED SUCCESSFULLY

有没有佬觉得可行？