• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    DeepSeek Releases Open-Source Agent Harness v0.1

    The open-source framework uses a plugin architecture for mixing AI components.

    SZ
    DE
    KA
    26 Sources, 49d ago, first seen 49d ago

    TLDR

    DeepSeek AI posted the developer preview of DeepSeek Harness v0.1. Multiple accounts described the release as open-source under an MIT license and built on the Cordis meta-framework. The tool provides an extensible agent runtime where models, tools, sandboxes, and orchestration pieces can be swapped via plugins. Visible replies noted its modular design and potential use for multi-agent web UIs, code, and bash tasks. One commenter called the move trolling while others expressed excitement over the public availability.

    Combined views

    4M

    26 Sources, first seen 49d ago

    Combined views

    4M

    26 Sources, first seen 49d ago

    24.9K likes
    24.9K likes
    1.2K comments
    8.4K saves
    2.5K reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    1.2K comments
    8.4K saves
    2.5K reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    26 Sources

    @teortaxesTexMore excited about DSH than V4 Pro now
    @AndrewCurran_@teortaxesTex Bloomberg reported about this.
    @deepseek_ai🧩 DeepSeek Harness v0.1 is now available in Developer Preview! 🔹 We’re opening it up to developers building agent harnesses worldwide and open-sourcing the codebase in MIT license. 🔹 Powered by the Cordis meta-framework, DeepSeek Harness is an agent harness built around one core idea: Everything is a plugin. Models, tools, skills, sessions, sandboxes, filesystems, loops, orchestration, and UI are ALL implemented as plugins, and can be mixed, matched, replaced, and extended. Try it now! https://github.com/deepseek-ai/deepseek-harness
    @tianyiDeepSeek Harness was just released with MIT license. The current 0.1.0 version is a developer preview, and may still have many rough edges. Feedback is welcome! DeepSeek Harness 已经以MIT协议开源发布。现在的0.1.0版本是一个面向Harness开发者的预览版，还很不完善。恳请大家多提提宝贵意见。 https://github.com/deepseek-ai/deepseek-harness https://deepseek.com/harness/
    @kimmonismusDeepSeek has open-sourced DeepSeek Harness, their framework for building and running AI agents. Digging into it now.
    @Dan_Jeffries1God why do we have no American companies with this kind of openeness and energy and spirit? It's just depressing. I really hope Meta's return to glory is it and that Reflection comes out swinging. We've just completely lost the energy and spirit of Linux (in general) and companies like Hashicorp, MongoDb and the old Red Hat.
    @thdxruhhh what does this mean
    @yacineMTBPlease, please god let it be kino and not slop
    @eliebakouchamazing release. it's a web UI with multiple harnesses inside it, you can spawn claude code and codex agent through their SDK "deepseek harness" supports different "modes" by default (which are harnesses): code mode with programmatic tool calling (in typescript), bash+edit (usually used in evals), or standard with write/read tool etc.. there is first class KV cache aware design, basically making sure the KV cache is never altered. for every change they don't modify previous history (prefix) but append something at the end that states the modification. i expect other harnesses to do the same but you never know? very interesting how this harness was developed too. it was HEAVILY agent first coded, there is an entire folder called ".agents/notes" that records every decision and proposal that agents left there (including some cool info like the fact that they also use it for post training) the UI is beautiful, especially the trajectory view. they have a full paper detailing composability and their system design for "plugins", this is going to be super helpful
    @Nag1ovo很荣幸入选参与了 Deepseek Harness 的内测！ 关于 DSH 相关技术本身，我没有什么太多想讲的，想必各大 AI 自媒体都会第一时间做出技术解析的流水线文章。我只是想聊聊这个温馨的技术群。 我很自豪自己在内测期间不曾对外部暗示或者泄露任何内测信息，并站好了内测的最后一班岗。 这段时间最大的收获，是认识了一群很有趣的群友：讨论、请教、互相拍砖，水群、语音、谈天说地。内测收官，工具很香，朋友更香。 内测期间，我花了点时间做了三个“重量级”插件，欢迎大家使用，好用/好玩的话给群主点个🌟吧！分别是： 1. https://github.com/Nagi-ovo/dsh-ads：系兄弟就来蹬我！青春记忆里的广告都回来了！ 2. https://github.com/Nagi-ovo/dsh-visualize：可交互的生成式 UI，可以像 Codex 里一样 ADHD 友好阅读模型的输出！ 3. https://github.com/Nagi-ovo/dsh-find-plugins：找到你最需要的那个 dsh 插件！ 关于 DeepSeek，不在你巅峰时慕名而来。作为从 V2 开始考到现在的老鲸考子，我会坚持考下去的！ 最后，乐队什么的，下次也一起组吧！🤗

    26 Sources

    @teortaxesTexMore excited about DSH than V4 Pro now
    @AndrewCurran_@teortaxesTex Bloomberg reported about this.
    @deepseek_ai🧩 DeepSeek Harness v0.1 is now available in Developer Preview! 🔹 We’re opening it up to developers building agent harnesses worldwide and open-sourcing the codebase in MIT license. 🔹 Powered by the Cordis meta-framework, DeepSeek Harness is an agent harness built around one core idea: Everything is a plugin. Models, tools, skills, sessions, sandboxes, filesystems, loops, orchestration, and UI are ALL implemented as plugins, and can be mixed, matched, replaced, and extended. Try it now! https://github.com/deepseek-ai/deepseek-harness
    @tianyiDeepSeek Harness was just released with MIT license. The current 0.1.0 version is a developer preview, and may still have many rough edges. Feedback is welcome! DeepSeek Harness 已经以MIT协议开源发布。现在的0.1.0版本是一个面向Harness开发者的预览版，还很不完善。恳请大家多提提宝贵意见。 https://github.com/deepseek-ai/deepseek-harness https://deepseek.com/harness/
    @kimmonismusDeepSeek has open-sourced DeepSeek Harness, their framework for building and running AI agents. Digging into it now.
    @Dan_Jeffries1God why do we have no American companies with this kind of openeness and energy and spirit? It's just depressing. I really hope Meta's return to glory is it and that Reflection comes out swinging. We've just completely lost the energy and spirit of Linux (in general) and companies like Hashicorp, MongoDb and the old Red Hat.
    @thdxruhhh what does this mean
    @yacineMTBPlease, please god let it be kino and not slop
    @eliebakouchamazing release. it's a web UI with multiple harnesses inside it, you can spawn claude code and codex agent through their SDK "deepseek harness" supports different "modes" by default (which are harnesses): code mode with programmatic tool calling (in typescript), bash+edit (usually used in evals), or standard with write/read tool etc.. there is first class KV cache aware design, basically making sure the KV cache is never altered. for every change they don't modify previous history (prefix) but append something at the end that states the modification. i expect other harnesses to do the same but you never know? very interesting how this harness was developed too. it was HEAVILY agent first coded, there is an entire folder called ".agents/notes" that records every decision and proposal that agents left there (including some cool info like the fact that they also use it for post training) the UI is beautiful, especially the trajectory view. they have a full paper detailing composability and their system design for "plugins", this is going to be super helpful
    @Nag1ovo很荣幸入选参与了 Deepseek Harness 的内测！ 关于 DSH 相关技术本身，我没有什么太多想讲的，想必各大 AI 自媒体都会第一时间做出技术解析的流水线文章。我只是想聊聊这个温馨的技术群。 我很自豪自己在内测期间不曾对外部暗示或者泄露任何内测信息，并站好了内测的最后一班岗。 这段时间最大的收获，是认识了一群很有趣的群友：讨论、请教、互相拍砖，水群、语音、谈天说地。内测收官，工具很香，朋友更香。 内测期间，我花了点时间做了三个“重量级”插件，欢迎大家使用，好用/好玩的话给群主点个🌟吧！分别是： 1. https://github.com/Nagi-ovo/dsh-ads：系兄弟就来蹬我！青春记忆里的广告都回来了！ 2. https://github.com/Nagi-ovo/dsh-visualize：可交互的生成式 UI，可以像 Codex 里一样 ADHD 友好阅读模型的输出！ 3. https://github.com/Nagi-ovo/dsh-find-plugins：找到你最需要的那个 dsh 插件！ 关于 DeepSeek，不在你巅峰时慕名而来。作为从 V2 开始考到现在的老鲸考子，我会坚持考下去的！ 最后，乐队什么的，下次也一起组吧！🤗