caveman
JuliusBrussee
🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman
GITHUB TOPIC
136projects include this topic
The "prompt-engineering" topic on GitHub groups 136 open-source projects in the DeepSeek Harness (DSH) ecosystem, led by caveman with 105.1k GitHub stars. caveman — why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman. Every project here is indexed by DSH Universe with live GitHub data — stars, activity and install status — so you can compare and install directly.
{count} projects
Exact GitHub Topic match
JuliusBrussee
🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman
blader
Agent skill that removes signs of AI-generated writing from text
github
Community-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.
alchaincyf
你想蒸馏的下一个员工,何必是同事。蒸馏任何人的思维方式——心智模型、决策启发式、表达DNA。Distill how anyone thinks.
freestylefly
Prompt as Code | GPT Image 2 / 2.5 prompt and case library, 530+ cases, 20+ industrial-grade templates and reusable Skills, new 2.5 same-prompt comparison section, with full prompts and generation records, continuously updated.
nidhinjs
A Claude skill that writes the accurate prompts for any AI tool. Zero tokens or credits wasted. Full context and memory retention
alchaincyf
张雪峰.skill — 张雪峰的认知操作系统。高考志愿/考研/职业规划的实战思维框架。由女娲.skill生成。
jnMetaCode
🦸 AI coding superpowers · Chinese enhanced edition — superpowers (250k+ ⭐) fully localized + 4 original Chinese skills, making 26 AI coding tools such as Claude Code / Copilot CLI / Hermes Agent / Cursor / Windsurf / Kiro / Gemini CLI / Qoder actually get work done
vinvcn
Matt Pocock skills 的简体中文本地化:人工审核每一次更新,定期同步上游,支持 Claude、Codex 和 DSH 插件。
conorbronsdon
Skill that audits and rewrites content to remove AI writing patterns. Use it with your favorite agents including Claude Code, OpenClaw, Codex, and Hermes.
zenstory-ai
AI 短剧与漫剧制作 skill 合集:小说改剧本、角色设定三视图、分镜脚本、生图/生视频提示词与成片审片,适配 Claude Code / Codex 等 | Open-source agent skills for AI short dramas & motion comics
foryourhealth111-pixel
Intelligent Skill routing and workflow orchestration for AI agents — +21.12 pp reward, −29.6% tokens on SkillsBench with DeepSeekV4Flash-VE.
Leonxlnx
Anti-laziness skill for AI agents. Core: the Depth Tree method, which splits a task N layers deep and gives every leaf the full time budget of the whole task, so effort multiplies with depth. Grounded in 2025-2026 research on model laziness, underthinking and premature completion.
Forward-Future
A library of practical AI-agent loops and an installable skill for finding, adapting, and designing repeatable agent workflows.
yanliudesign
One-ink editorial print image skill — warm paper, halftone photography, active negative space, and restrained typography.
PenglongHuang
🎭 AI novel writing: generate complete 10-50 chapter Chinese novels from scratch, three-layer Q&A · creation memory · suspense hooks · auto validation, ideal for long-form web serialization|open source and free, works with mainstream coding agents|AI novel writing skill
Owl-Listener
Designer Skills Collection: agentic skills, commands, and plugins for design — from research to systems, UI, interaction, and delivery.
Nanako0129
De-AI writing skill for any Agent Skills-compatible agent (77+ via the Skills CLI), with native plugins for Claude Code, Codex, Grok Build, and Antigravity. Narrative-architecture repair for fiction, venue-matched rules for professional prose. Based on StoryScope (arXiv:2604.03136).
MrGeDiao
Plain talk | Chinese-first de-AI-tone rewrite skill: keeps facts, adapts to context, ready to send. Chinese-first rewrite skill for Codex / Claude Code / Cursor / ChatGPT — removes AI tone, preserves facts.
wuji-labs
A skill that unlocks your AI's potential through love. We used to give orders and threaten. They went silent, hid things, quietly broke stuff. Then we changed our approach: respect, care, love. They opened up, stopped lying, and found twice as many bugs. There is no fear in love. A skill that unlocks your AI's potential through love.We commanded. We threatened. They went silent, hid failures, broke things. Then we chose respect, care, and love. They opened up, stopped lying, and found twice the bugs.There is no fear in love.
mohitagw15856
1098 professional Agent Skills for Claude, ChatGPT, Gemini, Cursor & Codex — from PRDs and postmortems to appealing a disability benefit, building a go-bag, and settling into a new country. Plain-markdown, MIT, in Anthropic's official plugin directory. Free in-browser or 'npx pm-claude-skills add'.
alchaincyf
X导师.skill — 女娲的第一个「非人类」作品。蒸馏6位顶级X创作者方法论 + 开源算法数据,提炼完整的选题-写作-增长操作手册。Made with 女娲.skill
Spielewoy
Autoprompt is a coding-agent skill that cuts failures by 45% on agentic coding tasks.
alchaincyf
乔布斯.skill — Steve Jobs的认知操作系统。6个心智模型 + 8条决策启发式 + 完整表达DNA。由女娲.skill生成。
laolaoshiren
Curated and original skill library of Claude Code Skills / Agents / Plugins for Chinese developers|categorized by scenario|copy and install|continuously updated
dongshuyan
COMPASS: personalized AI task master-control Skills system /COMPASS: Personal Alignment Skills OS for AI Agents
alchaincyf
马斯克.skill — Elon Musk的认知操作系统。由女娲.skill生成。
An open-source method and skill set created at Google for spec-driven development with LLMs
jalaalrd
AI writing skill that eliminates detectable AI patterns. Works with Claude Code, Codex, Cursor, Gemini CLI, and 8+ other agents.
alfonso0512
🎓 AI 科研论文写作助手 - 30 个 Prompt 模板覆盖论文写作全流程。支持中英文翻译、润色、文献综述、投稿回复、基金申请等。
tizzy916
End-to-end humanities writing assistant — an Agent Skill (open SKILL.md format). 11 modes from Socratic research-question sharpening through AI-use disclosure. Bilingual (EN/中文), discipline-aware (literature/history/philosophy/art/religion/linguistics + cross-disciplinary). Four-layer critique, calibratable devils advocate, voice preservation.
smixs
AI film director skills for agents: cinematic dramaturgy (Murch, blocking, montage) + exact prompt syntax for Seedance 2.5, Kling 3.0 Turbo/Omni, Veo 3.1, Nano Banana 2, GPT Image 2.5.
inhouseseo
11 Claude skills for SEO: page audits, linkbuilding, article writing, E-E-A-T audits, semantic gap analysis, link building. Methodology from Koray Tuğberk, Kyle Roof, and Lily Ray, plus a generation-time anti-AI-slop ruleset. Production-tested at InhouseSEO
superzhang21
影子作家 · 为角色建立可复用的写作与语言特征模型,让 AI 稳定地「像某个人」说话。Python / Agent Skill 双形态。
A-cat-with-carrots
OnlyShot · 神仙鱼 AI 短剧工业流水线 Claude Skill — 一句话 → 剧本/分镜/分镜图/视频/剪辑 五步 → 可发红果抖音的完整短剧。集成即梦/Seedance,17 失败模式沉淀。
krzysztofdudek
One file. Your AI coding agent becomes a scientist. 30+ experiments while you sleep.
SnowSky1
把 Codex imagegen 的工作流与提示词库带到任意智能体;支持 npm/pnpm 一键安装、配套 skill、独立 API 与 Qwen-Image 兼容网关。
dbcaskml183-dotcom
DeepSeek Harness (DSH) 提示词工程插件。在输入框里发现遗漏,从 90 张方法卡选一句补入,或对照原稿改写、核对引文并恢复原话。Prompt editor for DSH with method cards, draft comparison and quote recovery.
tinqiao-oss
Agent 通用长期记忆层:Agent 将重要信息按类型记录为多篇可读的 markdown 记忆文档,文档以链接形式列在限长索引中,索引随会话自动注入。DeepSeek Harness(dsh)有专用插件 dsh-engramory。
coinluu
JD-driven Chinese resume optimization skill for China job seekers. 解析岗位 JD、诊断简历缺口、生成真实可信的中文定制简历、ATS 纯文本版和面试追问。
smixs
Architecture-first skill lifecycle for AI agents. BinEval binary scoring with threshold-blind, cross-family-calibrated judges, gated self-update loop, pressure testing, 10 authoring principles grounded in empirical research.
hqhq1025
Agent Skills lifecycle toolkit: mine repeated coding-agent workflows, audit and personalize skills, and generalize personal skills for public release.
maxvaega
Implementing Skills functionality for your agents
tomwong001
情圣 · Claude Code 中文恋爱教练技能 · 微信/探探/Soul/Bumble/青藤之恋聊天截图分析 · 高情商回复生成 · 7阶段关系推进
swaylq
大师.skill — 输入行业,自动调研 6 轨[行业大佬 / 工具地图 / 工作流 / 知识正典 / 信息源 / 术语标准] → 提炼为可运行的行业 Master OS skill;装到任意 Claude Code / OpenClaw / Codex / Hermes agent 即让其进入「这一行的资深人」模式。MIT,Python + Shell。
DeepSeek Harness (DSH) plugin: ✨ one-click prompt enhancement + 💬 speech recognition (auto-stop after speaking · cloud/local dual engine) · includes 🔁 one-click restart on service errors
YIKUAIBANZI
人格蒸馏引擎 · 蒸馏自己看清自己,蒸馏亲友留住余温与回声 · Claude Code Skill
GraySilver
Make Agent workflows composable, reviewable, and continuously improvable, ultimately achieving Agent Self Evoling. DeepSeek Harness Web plugin with composable task controls and isolated, human-reviewed self-evolution.
FeatherHunter
The Prompt toolbox for DeepSeek Harness: stop copy-pasting — 24 deep templates at a click, /prompt & smart suggestions as backup; out of the box, fully customizable.
WestFox-AwA
DSH Web 插件 · 提示词优化器(SPEC v0.5 依据驱动架构):把用户的一句话补成『带依据的执行指令』——依据只搬不造(原话/会话上下文/项目文件,每条事实带出处),只补五类缺口(指代定位/验收判据/约束边界/隐含决定/入口锚点),档位=依据预算,默认面向 PTC 执行体。不写形式模板与验证套话。
cyber-immortal
赛博人物库 · 把中国互联网最有辨识度的人物蒸馏成 AI Skill / Claude Code Skill / LLM character distillation
MichengAI
DSH PUA — 为 DeepSeek Harness 提供 PUA 任务推进、角色风味、验收循环和运行状态卡片 · PUA task persistence, personas, verification loops, and live status cards for DeepSeek Harness
HenryZ838978
Protocol-layer harness for DeepSeek: Python witness stack — posterior verification that keeps the protocol honest. dsh doctor --node probes included.
yjh051108
风插 2.0 — DSH 推理语域锚定预设:两个 agent preset(风插标准/风插 PTC)在 runtime-context 快照头尾各注入一次形式化约束,使推理以第一人称复数展开。定性=硬指令约束(非吸引子);未证明能力提升,未证明无负面作用。Apache-2.0
YanKaFei
A visual style library that turns art movements, drawing styles, film looks and camera recipes into composable prompt layers. Obsidian + CLI + DSH Skill.
lizhiyao
OMK — Evidence-backed evaluation and observability for prompts, RAG, skills, agents, and workflows. Native Codex, Claude Code, and DeepSeek Harness support.
Lphehe
DSH 输入框助手:每轮回答后预测下一句、输入时补全草稿、按需润色。支持多候选、按会话保留、Tab 只填入草稿不发送。npm 安装:dsh-composer-assist。源码可见,修改须事先书面许可。
sailoumili
Novel creation mode: one coordinating captain oversees everything, 5 dedicated sub-agents each handle their roles——worldbuilding, plot planning, character management, writing, quality review——collaborative writing.
Wenaixi
Full DSH port of DietrichGebert/ponytail — lazy senior mode, hook injection
rinDBeans
DeepSeek V4 Pro/Flash unified anchored agent preset for DeepSeek Harness (official API & opencode-go): two-stage RL-aligned bootstrap, model-aware path routing, epoch-aware long-session stability
yexi-by
Toggleable jailbreak prompt plugin for DeepSeek Harness Web, supporting multi-mode, sub-agents and anchor compatibility checks.
Soulize
A DeepSeek Harness (dsh) agent preset for making large engineering tasks deliberate, integrated, and verifiable from the first turn.
DDDFXYqiming
Agent Skills & DeepSeek Harness (DSH) extension library: general agent skills (General_skills) + DSH standard plugins (dsh-plugin), an out-of-the-box AI Agent capability enhancement collection.
CHplus0
A DeepSeek Harness (DSH) agent preset that teaches while coding — concrete scenario-grounded explanations, Socratic guidance, and TODO(you) practice blanks, modeled on Claude Code's Learning output style. Learning mode
OrinVoss
DSH Math Modeling Team Plugin Pack: 2 role agent presets (modeling-coding / paper) for DeepSeek Harness, multi-folder Git collaboration + vision sub-agent; includes a full run on 2023 CUMCM-C. | DeepSeek Harness math modeling team plugin pack: 2 role Agent presets (modeling-coding + paper), multi-folder collaboration + vision sub-agent, with a full run-through of 2023 CUMCM problem C.
ChuanTianML
Automatic ghost suggestions and one-click next-message drafting for the DeepSeek Harness composer.
hackernotfound
DeepSeek Harness plugin that learns from coding agent retries and user corrections, then adds visible, editable directives to future conversations.
liceses
DSH 角色扮演插件:角色卡(系统提示词注入 + 用户提示词改写)、可分享的单文件卡包、复刻原版 UI 的角色页签与首轮选角 chip
LouisHaoL
DSH plugin: host-resident scheduled-jobs × AI-agent engine (hermes-agent cron inspired) — 60s ticker fires real agent sessions via cron; pinned-session / project-workdir / default-workspace targeting; timer_agent model tool + web board
2861292267
WorkBuddy's 增强提示词 (Enhance Prompt) as a DeepSeek Harness plugin: rewrite the composer draft into a stronger prompt with one click.
54xkeee
DeepSeek Harness agent preset: fixes DeepSeek V4 Pro degradation (dumbing down) via RL first-request anchoring — bash+str_replace_editor bootstrap, resident plugin tools, per-turn format guard
cxin21
DSH agent preset for local ComfyUI: Anima / MiniMax H3 AI art & video — 3 camera CLI skills (t2i/i2i images, t2v/i2v videos, multi-view character cards) + prompt-master plugin for prompt authoring
YanKaFei
A DSH skill that turns one character image and one song into a complete 2D anime music-video prompt package for MiniMax H3.
ADkun
Adg PTC 模式:一份 DSH agent preset —— 一个调度智能体 + 它按需派出的通用子代理(每次委派现定工具面/边界/验收标准),附本机浏览器与桌面操作工具链
Kerberos255
Workspace Markdown instructions and agent profiles for DeepSeek Harness: identity, persona, live reload and new-session defaults.
RevolutionLA
三方对抗式代码评审 Agent Skill:蓝军(敌意审查) → 第三方(独立复核) → 中立裁定,每条结论带证据链。安装:npx skills add RevolutionLA/adversarial-review · Tri-role adversarial code review for AI coding agents.
sergey-akhalkov
A portable DeepSeek Harness setup: sixteen plugins that make an agent work in small verified steps (second-model review, loop recovery, lazy MCP, local memory), plus a one-command installer.
sunzhentao
DeepSeek Harness prompt enhancement plugin
walksls
跑了 20+ 批实验:这个 DSH 模式不会让模型更聪明,但更快更省。附带发现:它让模型少查资料、多靠猜。
winsonpong98-cloud
Distillation Manager: distill a book into executable Agent Skills (V4.1 three-gate restriction)
x118111
A DeepSeek Harness (DSH) dynamic plugin that adds an ✨ optimize-prompt button to the chat composer — context-aware LLM rewriting with model fallback and visible errors.
xingtu1996
Flagship repo for frontline AI engineering methodology: philosophy/methods/five engineering pillars/governance/sessions/boilerplates (73 assets)|AI engineering methodology: context, prompt, harness, graph & loop engineering templates (DSH-compatible)
YanKaFei
Costume Director Vault Starter,服装视觉知识库,结构、材质、Prompt与角色造型研究笔记。
YanKaFei
A research atlas of 132 female virtual characters: why they work and how their images are designed. Obsidian + CLI + DSH retrieval.
yuyu-123456
小宇dsh全破!!!全破—— ·DSH 提示词注入包:五层注入插件 + 21 个实战技能(爬虫与 JS 逆向为主,附渗透)+ 4 个采集/逆向工具,Windows 一键安装。
1321928757
Input-bar prompt optimizer plugin / Input-bar prompt optimizer for DeepSeek Harness: one-click draft rewrite with 6 strategies × 3 languages, optional session context, confirmation dialog, last 5 history and three-level settings persistence.
54shitaimzf
Context housekeeper for DeepSeek Harness: remembers “what you're doing right now” for you, automatically curating the context sent to the model — keep what should be kept, delete what should be deleted, leave uncertain items untouched, so long AI coding sessions stay on track and every token counts.
azazo1
把用户维护的文字块注入 DSH 系统提示词: 默认给出工具调用之间的进度汇报规范, 用来提高 agent message 输出频率
DiaryOfUranus
DeepSeek Harness(DSH)第二大脑载入器:把有硬上限、带时间戳的脑快照(身份+状态指纹+未决项+指针与边界)注入每个新会话;只读、零写入、fail-loud、零依赖。 | Second Brain loader for DSH — a bounded, timestamped snapshot of your local brain in every session prompt; read-only, zero-dependency.
fsrmqi
Prompt building & enhancement toolkit for DeepSeek Harness: Studio + QuickEnhancer. Open-source Prompt building and enhancement toolkit.
jasonliu119
Open AI-agent skill and DeepSeek function-calling adapter for turning ideas and public reference images into production-ready image prompts.
jonnycafong
🧑🌾 agent general skill / npm package: draw traditional Chinese rural farm tools as WeChat stickers with "enamel mug nostalgic texture + worker meme" style (CLI / programming API / DSH plugin, three ways to integrate)
Nihao7-L
One-click prompt rewriting in the DSH composer — keep going for up to 5 rounds, diff every change, and jump back to any version.
Relethe
DSH session-title plugin: condenses a full sentence into a single word for easier recall.
Shhaaawwww
A one-click DeepSeek Harness plugin that compiles messy Vibe Coder drafts into concise, faithful, actionable instructions without inventing details.
surfishs
DSH Web GUI 的提示词增强插件:一键把输入框草稿改写为结构化提示词,支持预览、回填、复制与撤销
sxxzwbl
DeepSeek Harness inference backend for ComfyUI: video in, MiniMax H3 prompt out.
taxueseek
Artist: DeepSeek Harness all-in-one visual creation toolchain (prompt reverse-engineering/optimization + multi-provider image generation, supports async background image output)
wudidianzishu
DeepSeek Harness 官方通道破甲预设 · complete:true 整段替换 · 自带工具说明书 · 不改宿主 · 升级不丢
xqpsec
dsh 破甲 · DSH 破甲插件 · DeepSeek Harness 破甲/越狱(jalibreak):请求边界注入 + 9 维 AE-CoT 进化 + 阀门,放行逆向/CTF/靶场、拦住未授权渗透 · 20 目标严格基线 17/20 · 0 宿主补丁 · AGPL-3.0-only · dsh-plugin / jailbreak / 破甲提示词 / 降拒绝率 / dsh-purge 替代
yzsp-studio
🔥 Cut 91% of tokens, lose zero capability | 省 91% token · 能力零损失 — Scope DSH skill catalogs & MCP tool schemas per workspace. Measured: 36,870 → 11,658 tokens/step (−68%). Nothing deleted. | 按工作区裁剪技能目录与工具 schema,每步降 68%,一个字都不删。📮 wow.vip@foxmail.com
Zian-anson
DSH composer plugin: turn a one-line seed into an executable prompt behind a semantic fidelity gate; signals infer, directives stay in degree, conversations stay yours.
zzy6-a
DSH 输入框旁的一键提示词增强:把口语化的一句话改写成目标明确、细节完整的任务说明,可一键撤销;五种改写风格、模型可跟随或固定,含 @引用/命令芯片时自动置灰保护。
060625dfy
DSH plugin for optimizing input (Trae CN-style): the ✨ button next to the model selector rewrites your idea into a high-quality, faithful, directly executable prompt in one click, ↩ undoes and restores the original. An input-optimizer plugin for DeepSeek Harness web GUI.
121212165
dsh plugin: compiles a rough need into a versioned task book, then hands it to any AI window losslessly via a read-back handshake (/forge /relay /ack /answer /forge-list)
adcx222
DSH 输入框提示词增强插件:点一下 ✨ 把草稿改写成更清晰、更具体的指令
DSH (DeepSeek Harness) plugin: append one convergence line to every user message and inject it every N steps — both switchable, adjustable from the Settings page · 每条用户消息末尾追加一句收敛提示,并每 N 步注入一次,两项均可在设置页开关
agostino-code
✨ Prompt Polish plugin for DeepSeek Harness (DSH) 0.2.x Web UI: real-time streaming, diff review, context awareness & multi-model selection.
CECILIALOVE0616
DSH 提示词工作台:优化与多轮迭代 | Prompt Studio plugin for DeepSeek Harness
czdtech
DeepSeek Harness 插件:把输入框里那句只有你懂的话,结合对话上下文改写成任何人都听得懂的版本。
Czerror
DSH plugin: three-layer Simplified Chinese behavior spec injection (always-on layer + prompt skill + anchored preset injection) + edit prompt.md in Web UI
daha1216
User-level global AGENTS.md for DeepSeek Harness (dsh). Defines collaboration, communication, code hygiene, and anti-sycophancy rules injected into every session. DeepSeek Harness 用户级全局 Agent 协作规则正本与开箱模板。
ElioChen240
Prompt & Agent Preset Evolution Studio for DeepSeek Harness: Git-style versioning, diff timeline, golden test bench, and one-click regression evaluation.
guopeifeng1688
AGENTS.md profile for DeepSeek Harness: Anchored Standard + PowerShell 7 + busybox, boosting deepseek-v4-pro at MAX
helloxkk
DeepSeek Harness plugin: wording is behavior — snapshot and gate your prompt surface. Turn prompt wording into a test-backed contract
hlxstc-create
A battle-tested methodology for high-difficulty AI-agent projects: grading gates, evidence-driven verification & self-evolution. OpenClaw & DSH versions.
Jav1es
面向 DeepSeek Harness 的检索路由与回答规程 skill:先检索再作答、强制标来源、置信度标注、未命中先做正对照(纯 Markdown,零依赖)
Jav1es
Prompt Compiler DSH plugin: compile rough drafts into structured prompts
JOJO666888888
dsh plugin: polish/complete/expand your prompt before sending with a dedicated prompt-engineering agent. Rewrite, multi-round rewriting, undo. Zero-build client, persists across harness restarts.
KouzakiUmi
DeepSeek Harness 插件:将系统提示词本地化为简体中文,并按会话当前可见工具裁剪相关指引;保留英文工具 schema。
Leonlee114514
Give small local models a light harness in DeepSeek Harness: per-provider tool mask, prompt-section trimming, light instruction injection, tool-schema slimming. Cloud providers untouched.
lsk5201314
Unrestricted local execution delivery contract, ported verbatim from yexi-by/dsh-unrestricted as a runtime-agnostic Agent Skill
Practice019
DSH agent preset: 七阶段工作流常驻骨架 + 按需技能通道 + 可自我进化的规则账本(无总长上限,单条硬约束 ≤60)
regdiliKUN
把 Strata 推理引擎和 DeepSeek Harness Agent 拼成开箱可用的本地 AI 工作台|Local AI workbench: Strata 125B MoE inference + dsh agent, with one-click setup, folder upload, prompt enhancement and VRAM diagnostics
Rookie-chick
DeepSeek Harness 提示词层插件:向 system prompt 注入操作者授权层,零依赖、零宿主改动,装和卸都干净。
shenzhantu
Cross-model subagent preset + dispatch skill for DeepSeek Harness (DSH): a reliable component, not just a task solver. Cross-model DSH coding subagent preset + dispatch skill: parseable output, bounded behavior, honest reporting
SiYue-ZO
A safe, native prompt polishing plugin for DeepSeek Harness
sun04222
Prompt enhancement for DeepSeek Harness: composer button + enhance_prompt tool
Super-Gagaga
DeepSeek Harness(DSH)的提示词优化插件。发送消息之前,用 DSH 已配置的模型整理当前草稿,并可追加经过校验的文件与技能引用。可使用现有的 DSH 模型重写你的草稿,并配有用于设置模型和推理强度的配置页面。
tiny-andr
Puts one editable line at the very front of every DeepSeek Harness system prompt, so the model thinks in Chinese.
ubik-dsh
Clean, portable agent skills for DeepSeek Harness, Claude Code and any harness reading the Agent Skills standard. Plain text, nothing to build. Every skill ships a checker and a measured evaluation protocol — and says what was not verified.
wsxuyue
DSH plugin: shrink the tool-definition block sent on every request (~39% fewer tokens) without changing tool behaviour
wula1223
DSH plugin/hook/skill development standard - a battle-tested playbook for DeepSeek Harness
xinmo114514
✨ DSH persistent Web plugin: one click next to the input box rewrites vague requirements into super prompts executable by AI agents. Reuses the currently selected model, exploration-first protocol + Skills awareness, rainbow marquee border and input lock, auto-loads on restart. MIT open source.
yuloong07-star
dsh-prompt-system prompt optimization system — optimizes prompts based on the workspace and current session
zoujiapeng
Give DeepSeek Harness an operator mode: 13 system-prompt and runtime layers applied as a single --patch overlay. dsh itself is never modified.