Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add techflag/workdsh --skill workdsh-workbench-devgit clone --depth 1 https://github.com/techflag/workdshWrote 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/techflag/workdsh/workdsh-workbench-dev)<a href="https://agentmods.dev/skills/techflag/workdsh/workdsh-workbench-dev"><img src="https://agentmods.dev/badge/skills/techflag/workdsh/workdsh-workbench-dev/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/techflag/workdsh/workdsh-workbench-dev"><img src="https://agentmods.dev/badge/skills/techflag/workdsh/workdsh-workbench-dev.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00034 | $0.01955 |
| Opus 5.5 | $0.00014 | $0.00782 |
| Sonnet 5.5 | $0.00007 | $0.00391 |
| Haiku 4.5 | $0.00003 | $0.00196 |
Grade A, and why
workdsh-workbench-dev 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 6d ago.
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 — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.
工作台插件开发
先通过宿主技能工具加载官方 cordis-plugin-development,在官方 Creator(创造模式,cordis)中使用原生会话、权限与工具链。本 Skill 只补充 WorkDSH 的业务约束;Host、Client、Loader、Slots、插件包和安装流程遵循官方 Skill,不另造开发 Agent、会话界面或安装器。
再阅读 需求定义 和 工作台产品体验,从一句想法开始时先阅读 需求梳理,首次制作阅读 首次开发,修改已有插件阅读 继续调整。用户本轮的范围、布局选择和暂停要求优先;这些文档不代表能力已经实现。
共同规则
- 从“和 AI 一起梳理”进入且尚未确认需求时,先调用
workbench_read_creation_request,阅读 需求梳理,帮助用户明确目标并保存可编辑摘要,不生成源码、构建或安装。用户在左侧确认后,当前会话会关联正式需求与开发工程;已进入开发时先调用workbench_read_development_context,读取当前原生会话实际关联的需求、源码、稳定插件身份、目标实例和候选包;不根据旧提示词猜测归属。先使用上下文sdk提供的已解析公开文档与类型;不要用Service.listService查询已知未收录的 WorkDSH 注册服务。官方 API 按官方 Skill 核对本轮用到的服务、事件、工具和 Slot,避免枚举无关接口。WorkDSH 扩展接口须按 SDK 核验 单独核对:官方 Service 目录未收录不等于服务未运行,Config 的absent不等于插件未安装。WorkDSH 业务 SDK 和模板也以宿主实际交付的版本为准。不得猜测接口、复制私有实现或伪造回执;缺能力时保留源码和具体未完成项。 - 先定义业务对象、状态、用户动作、AI 分工和成果落点,再安排模块。项目关联是可选的;项目、资料、技能、专家、连接器使用各自公开服务和授权,不复制管理系统。一个包只声明一个工作台,后续保持包名与工作台 ID;本地制作不要求公开 GitHub 仓库。
- 保持用户已选布局,包括业务在左、原生会话在右的并排方式;不因套用模板强制重构。复用完整原生会话及其执行过程,不自制 Composer、聊天进度或工具轨迹。AI 通过真实业务工具读取当前对象、执行并写回候选或独立草稿,页面读取后供用户继续加工或采用;跳转聊天、复制提示词或回复结束都不等于完成业务协作。
- Excel、CSV 等文件复用系统 Office 的公开预览、编辑、保存和导出能力,不另写电子表格编辑器。业务页面负责字段对应、核对和统计;资料引用准确的资产及版本,统计由代码对完整已授权输入的固定修订确定性计算,不能拿截断预览作统计源。保存新文件版本后明确采用,历史分析与报告保留原快照。
- 对象、输入版本、会话和身份必须匹配;切换对象、账号或迭代后,旧回复不能写入新对象。保持现有源码、会话历史、用户草稿和业务数据,交接不能覆盖未发送的输入。企业请求只走 Host 已授权业务服务,不向页面提供通用带凭据代理;token、密码和密钥不进入页面、提示词、源码、日志、导出或插件包。
- 工作台默认按可持续使用的软件交付:围绕业务记录组织首页、列表、详情与返回路径,复杂任务使用应用内导航或页签,不把全部功能纵向堆成无限长页;多记录和页面数量根据本轮业务决定,用户明确要求的一次性工具除外。具体执行 软件形态要求。页面使用中文业务字段,说明当前任务、结果与下一步,呈现正常、异常、空数据和失败恢复状态。业务插件不自行增加宿主一级菜单;平台统一通过原生“工作台”入口提供已安装应用及开发记录,不生成平级菜单。只使用实际已声明的官方公开插槽;不修改官方侧栏包、注入私有 DOM 或用缩进伪造菜单分组。平台提供的返回工作台和常用应用设置不在业务页重复实现;业务内部仍需提供返回模块或记录列表的路径,不能用浏览器返回代替。具体记录直达需要稳定对象标识和访问核对,不能拿会话链接冒充。局部修改沿用同一源码和插件身份。
- 通用安装使用官方
plugin_manager install_bundle,以已核实 tgz 绝对路径、enabled:false安装;再用候选 version、sha256 调用workbench_verify_native_installation核对字节并登记。activationVerified:false只表示本工具未核验启用,绝不表示停用或启用失败,不因此重复安装。用户已授权启用或打开试用时,按回执list_bundles(沿 nextOffset 分页)找到实际包名并核对版本、error,再set_bundle,target 为该包名、enabled 为 true;读取 application、warnings、nextStep。不能猜 entryId、拼 include: 前缀或用set_plugin代替 bundle 选择;不能手改 Profile。不具备启用授权时,由用户在工作台列表启用。安装结果未知先核对事实;restart-required 如实报告,不自行重启其他任务。安装、登记、活动版本和页面可用分别报告。 - 依据用户本轮授权推进。要求先讨论时先明确工作路径;要求看效果时提供可交互效果并说明模拟范围;暂停打包时不自动构建或安装。生成代码、包检查通过、安装成功和页面可用分别报告;构建与 Slot 注册不能代替真实服务和用户操作验收。
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago First seen · 36 lines · 34 tokens per session scan A faf4bfd81985
workdsh-workbench-dev is a skill published in the GitHub repository techflag/workdsh (265 stars, last pushed 6d ago), licensed MIT. It adds 34 tokens to every session and 1,955 once invoked, about $0.0001 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 2026-10-18.
Other skills, from other repositories
modsearch
Plug-in web search, X (Twitter) search, and page fetch for models without native web access. Use whenever the task needs current information, external facts, source links, posts from X, or the content of a specific URL, and the active model/harness has no native search or fetch tool. Runs the modsearch CLI to return…
acryl-build-plugin
Use when the user asks you to add a feature, tool, panel, command or integration to ACRYL itself (build a plugin or extension). The end-to-end workflow, from reading the docs to a live plugin.
acryl-share-plugin
Use when the user wants to save, share, publish, list publicly or version a plugin, or pull one from GitHub, npm, a file or the plugin store.
manage-taskboard
Manage work in the native DeepSeek Harness Taskboard with exact task ids and optimistic versions. Use when an Agent must inspect project work, claim an eligible todo, record progress or blockers, verify an implementation, submit it for human review, or release its own claim; also use when a human asks how to accept…
acryl-add-ui
Use when an ACRYL extension needs UI in the Web or Desktop app (a button, panel, sidebar tab, settings section, dashboard).
acryl-change-plugin
Use when the user asks you to change, fix, improve, restyle, extend or remove an extension or plugin that already exists in ACRYL.