Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/znlgis/opengis-skillsnpx agentmods add skills/znlgis/opengis-skills/deepseek-harnessWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/znlgis/opengis-skills/deepseek-harness)<a href="https://agentmods.dev/skills/znlgis/opengis-skills/deepseek-harness"><img src="https://agentmods.dev/badge/skills/znlgis/opengis-skills/deepseek-harness/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/znlgis/opengis-skills/deepseek-harness"><img src="https://agentmods.dev/badge/skills/znlgis/opengis-skills/deepseek-harness.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 8 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 46 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 74 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 77 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 84 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 87 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 90 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 132 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 221 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00119 | $0.04067 |
| Opus 5.5 | $0.00048 | $0.01627 |
| Sonnet 5.5 | $0.00024 | $0.00813 |
| Haiku 4.5 | $0.00012 | $0.00407 |
Grade A, and why
deepseek-harness scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 315 lines — stays where its author put it; the contents beside it link to each section on GitHub.
项目地址： https://github.com/deepseek-ai/deepseek-harness
官方文档： https://deepseek-harness.github.io/deepseek-harness/
npm：
@deepseek-ai/dsh（0.2.1-rc.2，截至 2026-10）｜ PyPI：deepseek-harness-sdk许可证： MIT
概述
DeepSeek Harness（命令 dsh）是 DeepSeek AI 开源的智能体框架，"一切皆插件"架构，底层是 Cordis（设计见论文 A Programming Paradigm for Spatiotemporal Composability）。一个 dsh 进程 = 一棵由 plugin bundle 逐层 patch 出来的插件树；Web / headless / SDK / ACP 都是不同的 profile，不是不同的可执行文件。
- profile 即入口：
dsh web、dsh --profile headless "job"、dsh --profile sdk、dsh --profile acp共用同一个 npm 包 - 分层组合：shipped bundle → profile 的
cordis.patch.yml→$DSH_HOME/cordis.patch.yml→--patch覆盖层 - 插件形态：TypeScript 模块导出
name/inject/apply(ctx)，工具用defineTool注册 - 双语言自动化：
dsh --profile headless一条命令出结果；Python 侧有deepseek-harness-sdk - developer preview：官方明确会有 breaking changes，锁版本使用
环境准备
方式一：npm 直接跑（最快验证）
# 需要 Node.js；首次会下载依赖
npx @deepseek-ai/dsh web
默认在 http://127.0.0.1:3080 起 Web UI 并尝试打开浏览器；SSH 场景只打印地址。加 --no-open 不自动开浏览器：
npx @deepseek-ai/dsh web --no-open
方式二：仓库源码构建（开发插件时用）
git clone https://github.com/deepseek-ai/deepseek-harness.git
cd deepseek-harness
pnpm install
pnpm run build
pnpm dsh web
方式三：Python SDK
python -m venv .venv
. .venv/bin/activate # Windows: .venv\Scripts\Activate.ps1
python -m pip install deepseek-harness-sdk
SDK 自带匹配的原生运行时 wheel 和 dsh 命令，跑 SDK 不需要系统 Node.js；只有用 dsh plugin 管理插件时才需要 pnpm。
验证：
dsh --version # 打印运行时版本
npx @deepseek-ai/dsh --help # 启动器自身的参数（不是 web app 的参数）
profile 启动时会校验插件声明的
engines.dsh版本区间，不兼容需要显式声明精确版本豁免。
核心用法
入口模式
| 命令 | 用途 |
|---|---|
dsh <name> / dsh --profile <name> |
启动 $DSH_HOME/profiles/<name> 下的 profile |
dsh web |
Web UI（默认 127.0.0.1:3080） |
dsh --profile headless "job" |
跑一个持久化会话，打印最终答案后退出 |
dsh --profile sdk / sdk-minimal |
通过 JSON-RPC stdio 服务 SDK 客户端 |
dsh --profile acp |
通过 ACP stdio 服务自动化客户端 |
dsh --profile <name> --from-default-profile <template> |
用 shipped 模板重建一个自定义 profile |
dsh plugin --profile <name> <pnpm args> |
转发给 pnpm 管理该 profile 的插件 |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday Changed · +71 lines · +51 tokens per session 72cd94602fdc
- 14d ago Changed · +1 lines 7e58c607f068
- 15d ago Changed · +1 lines 9e855db0e1b4
- 1mo ago Changed cc41cbe8dab1
- 1mo ago Changed 2d44984107c6
- 2mo ago First seen · 242 lines · 68 tokens per session scan A 5b1892730a97
deepseek-harness is a skill published in the GitHub repository znlgis/opengis-skills (75 stars, last pushed 2d ago), licensed MIT. It adds 119 tokens to every session and 4,067 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
journal
A mirror for emotional truth. Use when the user wants to journal, reflect, or "tell me the truth about how I feel." Goes for the feeling beneath the event, names avoidance, and refuses easy answers — not a cheerful prompt list.
deepseek-harness
Use this skill whenever the user wants to call DeepSeek V4-Pro / V4-Flash (or its legacy aliases deepseek-chat / deepseek-reasoner), or you see code that imports from openai import OpenAI with baseurl="https://api.deepseek.com". This skill teaches you the 10 protocol contract rules required to avoid the 16 documented…
tg
Use when a task needs Telegram reads, searches, raw TL requests, media, or sends through the tg runtime.
visionary-cli
Analyze images with DeepSeek's vision model via the visionary-server CLI. Use this whenever the user provides an image, photo, screenshot, or document with images - run vision to look at it rather than guessing.
cuml-machine-learning
Use for GPU-accelerated machine learning on tabular data using NVIDIA cuML. Triggers when tasks involve classification, regression, clustering, dimensionality reduction, or model training on datasets.
tool-prompt-optimization
Optimize the description prompts an AI agent reads to learn its built-in tools (the .md files under prompts/tools/). Two halves: (1) measure how much of a prompt is already inferable from the tool's JSON parameter schema + name, to prune redundancy with evidence; (2) house authoring rules for what belongs in a tool…