Install the "develop-wework-plugin" agent skill from https://github.com/wecode-ai/Wegent/tree/main/wework/dsh/plugin-developer/codex-plugin/skills/develop-wework-plugin into .claude/skills/develop-wework-plugin/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "develop-wework-plugin", then confirm the skill loads.
Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
skills CLI
$ npx skills add wecode-ai/Wegent --skill develop-wework-plugin -a codex
Project install goes to .agents/skills/; add -g for ~/.codex/skills/.
Install the "develop-wework-plugin" agent skill from https://github.com/wecode-ai/Wegent/tree/main/wework/dsh/plugin-developer/codex-plugin/skills/develop-wework-plugin into .agents/skills/develop-wework-plugin/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "develop-wework-plugin", then confirm the skill loads.
Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add wecode-ai/Wegent --skill develop-wework-plugin -a cursor
Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.
Install the "develop-wework-plugin" agent skill from https://github.com/wecode-ai/Wegent/tree/main/wework/dsh/plugin-developer/codex-plugin/skills/develop-wework-plugin into .cursor/skills/develop-wework-plugin/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "develop-wework-plugin", then confirm the skill loads.
Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
skills CLI
$ npx skills add wecode-ai/Wegent --skill develop-wework-plugin -a gemini-cli
Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.
Install the "develop-wework-plugin" agent skill from https://github.com/wecode-ai/Wegent/tree/main/wework/dsh/plugin-developer/codex-plugin/skills/develop-wework-plugin into .gemini/skills/develop-wework-plugin/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "develop-wework-plugin", then confirm the skill loads.
Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
skills CLI
$ npx skills add wecode-ai/Wegent --skill develop-wework-plugin -a github-copilot
Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.
Install the "develop-wework-plugin" agent skill from https://github.com/wecode-ai/Wegent/tree/main/wework/dsh/plugin-developer/codex-plugin/skills/develop-wework-plugin into .github/skills/develop-wework-plugin/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "develop-wework-plugin", then confirm the skill loads.
GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add wecode-ai/Wegent --skill develop-wework-plugin -a opencode
OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
Install the "develop-wework-plugin" agent skill from https://github.com/wecode-ai/Wegent/tree/main/wework/dsh/plugin-developer/codex-plugin/skills/develop-wework-plugin into .opencode/skills/develop-wework-plugin/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "develop-wework-plugin", then confirm the skill loads.
OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
Skill name
develop-wework-plugin
GitHub stars
868
Token cost
~3.5k tokens
SKILL.md length
1,498 words
Files
55 (incl. references, assets)
Skills in repo
8
Repo updated
First seen
Licence
Apache-2.0
At a glance
Create, extend, and debug Wework Core DSH plugins and their optional nested Codex plugins.
Works in 5 steps: Read .wework/plugin-development.json,… → When wework.codexPlugin is present, read… → Read every declared host, browser, or… → …
The user asks to develop a Wework plugin
SKILL.md covers Use the bundled development kit, Understand the package…, Inspect before editing and Choose the smallest extension…, plus 4 more sections
Runs JavaScript scripts from its folder
What it does
Develop Wework Plugin is an agent skill from wecode-ai/Wegent. Create, extend, and debug Wework Core DSH plugins and their optional nested Codex plugins. Use when the user asks to develop a Wework plugin, add a Wework UI extension, create a Skill inside that plugin, or diagnose it in the isolated plugin-development instance.
Its SKILL.md is about 3.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 59 other files, including reference files and assets (for example `assets/reference-plugins/README.md`, `assets/reference-plugins/endpoint-watch-demo/README.md` and `assets/reference-plugins/endpoint-watch-demo/client.js`).
It sits in Agent Workflows, covering Skill authoring and Hooks and plugins. The repository describes itself as: Plan, build, and deliver with an open-source, self-hostable AI workspace for coding, collaboration, and automation. The licence is Apache-2.0.
When your agent uses it
The user asks to develop a Wework plugin
Add a Wework UI extension
Create a Skill inside that plugin
Diagnose it in the isolated plugin-development instance
Example prompts
“/develop-wework-plugin”
Requirements
Node.js
Workflow steps
5 steps, taken from the first numbered list in SKILL.md.
1Read .wework/plugin-development.json, package.json, and the declared
2When wework.codexPlugin is present, read its .codex-plugin/plugin.json
3Read every declared host, browser, or sidecar entry affected by the change.
4Reuse public Wework DSH slots and services. Do not import private Wework
5Remove duplicate loaders, watchers, compatibility paths, and generated
What it can do on your machine
Read from SKILL.md and the folder at commit 428f207. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships script files (JavaScript, from the files we listed), which the agent can run.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Develop Wework Plugin loads about 3.5k tokens when it runs, and up to ~9.3k if it reads all its reference files. Until then it costs about 71 tokens; SKILL.md has 1,498 words of instructions outside code blocks.
Always· name and description, kept in context so the agent knows when to use it
~71
When it runs· the whole SKILL.md, loaded when a task matches
~3.5k
With references· SKILL.md plus every file in references/, read only if the agent opens them
~9.3k
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
Download SKILL.mdSave it as .claude/skills/develop-wework-plugin/SKILL.md (or your agent's skills folder). This skill also uses 54 other files; get the full folder from GitHub.
name
develop-wework-plugin
description
Create, extend, and debug Wework Core DSH plugins and their optional nested Codex plugins. Use when the user asks to develop a Wework plugin, add a Wework UI extension, create a Skill inside that plugin, or diagnose it in the isolated plugin-development instance.
Develop a Wework plugin
Treat the Wework package as the outer delivery unit. It may declare a nested
Codex plugin through wework.codexPlugin; the Codex plugin must never own or
gate the Wework plugin.
Use the bundled development kit
This Skill includes the complete public Wework extension catalog, one
exhaustive smoke-test plugin, three focused API references, and three
product-oriented showcase plugins:
Read references/extension-points.md before
choosing a UI surface. It documents descriptor fields, component props, and
the browser-module boundary.
Inspect assets/reference-plugins for runnable,
independently installable examples of Composer augmentation, persistent
workflow UI, website navigation, and typed desktop-host integration. For a
left-navigation button that opens a website in the Wework built-in browser,
start from sidebar-browser-panel-demo. Copy only the closest example into
the user's writable project, then delete every contribution the plugin does
not need.
Start with assets/showcase-plugins when the user
wants a product-quality example. Workspace Copilot, Test Explorer, and Dev
Environments represent popular AI-assistant, testing, and
development-environment categories with deliberately different interaction
models.
Never edit files inside an installed plugin cache. Resolve these resources
relative to this Skill, and copy examples to the active project first.
The catalog covers all public host slots:
Navigation and apps: wework.action, wework.app, wework.route,
wework.sidebar.navigation, wework.settings.page, and
wework.settings.section.
Project and workspace: wework.plugins.action,
wework.project.create.section, wework.project.work.section,
wework.workspace.menu.section, wework.workspace.tab,
wework.workspace.sidebar.tab, wework.workspace.toolbar.action,
wework.workspace.bottom-panel.tab, and
wework.runtime-profile.workspace-policy.
Composer: wework.composer.action.
Home and context: wework.home, wework.task.status,
wework.conversation.summary, and wework.board.card.status.
Shell: wework.shell.before, wework.shell.after, and
wework.shell.overlay.
It also documents the public ctx.wework.host, backend, commands,
composer, contributions, chat, testing, environments, context,
menus, keybindings, localization, configuration, storage, and
secrets services. Prefer a command-backed menu contribution when the host
already owns the button surface; use a raw UI slot when the plugin needs custom
rendering.
Wework creates .wework/plugin-development.json beside the project files to
classify the local project and enable the 插件调试 tab. Treat that file as
Wework project metadata, not as part of the distributable Codex plugin.
The boundaries are strict:
package.json, cordis.patch.yml, index.js, and client.js belong to the
outer Wework package.
New projects contain only the outer Wework package by default. Add
wework.codexPlugin and a nested Codex plugin only when the requested product
capability actually includes Codex Skills, MCP servers, agents, or other
official Codex plugin resources.
A nested Codex plugin uses the official .codex-plugin/plugin.json format
and official Codex folders only. Do not add Wework-only keys or files to its
manifest.
index.js runs in the Core DSH Node host. client.js runs in the browser
client. Share data through declared services or messages, not accidental
module state or startup order.
Third-party plugins consume public Wework DSH slots and services. Never
import private Wework application modules.
Inspect before editing
Read .wework/plugin-development.json, package.json, and the declared
dsh.bundle.patch.
When wework.codexPlugin is present, read its .codex-plugin/plugin.json
and every affected Skill or MCP entry below that directory.
Read every declared host, browser, or sidecar entry affected by the change.
Reuse public Wework DSH slots and services. Do not import private Wework
application modules from a user plugin.
Remove duplicate loaders, watchers, compatibility paths, and generated
examples that no longer serve the plugin.
Choose the smallest extension surface
Start from the user-visible outcome, then select the narrowest matching slot:
Add a complete application with wework.app.
Register the full-content route rendered by an existing workspace tab with
wework.route. It does not create a tab; when the tab's contentRoute
matches, the route component replaces that tab's entire content surface.
Add left navigation with wework.sidebar.navigation.
Create a selectable top-level workspace tab with wework.workspace.tab.
Add project-scoped inspection or controls with
wework.workspace.sidebar.tab.
Add compact workspace-level actions with
wework.workspace.toolbar.action.
Add a full bottom-panel tool with wework.workspace.bottom-panel.tab.
Add an input-adjacent action with wework.composer.action.
Add plugin-management actions with wework.plugins.action.
Add a standalone settings page with wework.settings.page.
Add controls to an existing settings page with wework.settings.section;
declare the target page in the contribution descriptor.
Replace the empty-task hero above the Composer with wework.home; use the
provided heading and call onSelectSuggestion to fill the active draft.
Use contextual and shell slots only when the UI genuinely belongs to that
lifecycle.
Do not build a parallel navigation, panel, or settings system when a public
slot already owns that surface.
When a left-navigation item should open a website beside the current workspace,
register a descriptor-only wework.workspace.sidebar.tab with
mode: 'iframe' and url, then point the navigation descriptor's
workspaceSidebarTab to that tab id. Do not register wework.app or create a
top-level workspace tab for this interaction.
Open a route in its own workspace tab
When a sidebar item must preserve the current tab and open the plugin route in
another top-level tab, put the public workspace-tab parameters in the
navigation descriptor's path:
Register the matching /example content with wework.route. A stable
workspaceTab value creates the tab on the first navigation and selects that
same tab later. Omitting workspaceTab intentionally replaces the active
tab's route. Use a generated ID only when every action must create another
tab. Do not import WorkspaceTabsContext, navigateTo, or other private
Wework modules into a plugin.
Show full SKILL.md (666 more words)Show less
Build the capability
Register reusable behavior once through ctx.wework.commands; invoke that
same command from menus, shortcuts, or plugin UI.
Register privileged Node behavior through the injected
ctx.weworkPluginRuntime, then call it from browser UI through a namespaced
ctx.wework.backend client. Do not invent plugin-specific loopback servers.
Use ctx.wework.host for typed desktop capabilities; do not call private
Electron bridge URLs.
Localize visible copy with ctx.wework.localization.translate(...). Standalone
plugins must not import Wework's private React localization hooks.
Publish state used by visibility and enablement rules through
ctx.wework.context.
Use ctx.wework.menus for standard Composer and workspace toolbar actions.
A command selected from composer.slash receives
invocation.composer; use its getValue, setValue, insertText, and
focus methods to edit the active draft without importing Composer internals.
Use ctx.wework.composer.references for resources that users should discover
and insert through the Composer @ menu.
Use ctx.wework.chat.providers, ctx.wework.testing.providers, or
ctx.wework.environments.providers when the capability must be reusable
across multiple host or plugin surfaces.
Use namespaced ctx.wework.storage for JSON state and
ctx.wework.secrets only for sensitive strings.
Declare settings through ctx.wework.configuration; do not invent a second
configuration store.
Register host-readable metadata through ctx.wework.contributions.
Register visual components directly through native ctx.slots.register.
Do not attach descriptors to component statics or flatten DSH kind, scope,
child slots, stores, and injected business faces into a generic UI API.
Inject the same slot with ctx.slots.inject so both registrations follow the
Core DSH lifecycle.
Keep contribution ids stable and provide data-testid values for interactive
controls.
Use the descriptor fields and component props documented for the selected
slot. Do not infer undocumented host internals.
Keep Node and browser dependencies in their declared entries. Browser code
must not call Node APIs directly.
When creating or changing a Skill, keep it inside the declared nested Codex
plugin, use official Codex structure, and keep its description
discriminating.
Keep the plugin cohesive: remove unused Demo contributions, generated
placeholders, duplicate watchers, and obsolete compatibility paths.
A minimal browser registration follows this shape:
Use the exhaustive Demo for API coverage and the reference plugins for
product-oriented patterns instead of expanding this snippet into guessed APIs.
Debug
Use the current project's right-side 插件调试 tab:
Validate the project and fix the structured errors it reports.
Start the isolated Wework development instance.
Exercise the exact UI surface contributed by the plugin.
Edit browser source and confirm the visible behavior changes through HMR.
Read host, browser, or sidecar logs when behavior does not match the source.
Restart Core DSH only after dependency, manifest, bundle-patch, host-process,
or other process-state changes that cannot be hot reloaded.
Operate Wework through the general wework desktop CLI. This is the required
control surface for inspecting and interacting with both the main Wework
instance and the isolated plugin-development instance:
--project . resolves the running Wework instance registered for the current
project, so commands do not contain cache directories, ports, tokens, process
ids, or other machine-specific values. Use wework desktop instances and an
explicit --instance only when more than one matching instance exists.
Use inspect before selecting a target, prefer stable data-testid selectors,
and verify every mutating command with wait or another inspect. The CLI
intentionally has no arbitrary JavaScript evaluation command. Do not bypass
the CLI with a plugin-owned control implementation.
Read the actual structured failure and logs before modifying code. Confirm the
changed behavior in the development instance; a file-write event alone is not
evidence that HMR succeeded.
The development instance has separate login state, data, executor home, Core
DSH profile, cache, and logs. Never copy authentication data from the main
instance.
Verify
Before handing off:
Validate the outer Wework package and its declared bundle patch.
Validate the nested official Codex plugin when present.
Run focused tests for changed host, browser, manifest, and Skill behavior.
Exercise both Node and browser halves when both exist.
Verify one real source edit through HMR.
Confirm install, enable, disable, and uninstall lifecycles when UI
contributions changed.
Confirm every interactive control has a stable data-testid.
Stop the isolated development instance and verify its temporary runtime
resources are cleaned up.
Develop Wework Plugin next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
Develop Wework Plugin compared with similar skills
A guide to creating and managing Claude Code skills with auto-activation: skill-rules.json triggers, hooks, enforcement levels, YAML frontmatter and progressive disclosure.
Explains how to write Claude Code slash commands: Markdown files with YAML frontmatter, arguments, file references, bash context and interactive prompts.
Shows how to build a Vibe plugin package in the Agent Plugins 1.0 format, with a plugin.json manifest and optional skills, MCP servers, hooks and other components.
Explains how to design and write skills for Claude Code plugins: SKILL.md frontmatter, bundled scripts, references and assets, and descriptions that trigger at the right time.
Validates an Agent Skill package against a portable contract with a Python checker, then picks the smallest set of instruction, capability or lifecycle primitives for the task.
Create or update a Wework Smart app based on DeepSeek Harness, including environment preparation, DSH plugin discovery, composition, built-in-browser verification, packaging, and local installation…
Create, extend, and debug Wework Core DSH plugins and their optional nested Codex plugins. Develop Wework Plugin is an agent skill from wecode-ai/Wegent. Create, extend, and debug Wework Core DSH plugins and their optional nested Codex plugins.
When should I use Develop Wework Plugin?
Develop Wework Plugin fits situations like: the user asks to develop a Wework plugin; add a Wework UI extension; create a Skill inside that plugin; diagnose it in the isolated plugin-development instance.
How do I install Develop Wework Plugin in Claude Code?
Run `npx skills add wecode-ai/Wegent --skill develop-wework-plugin -a claude-code`. Or copy the skill folder (wework/dsh/plugin-developer/codex-plugin/skills/develop-wework-plugin in wecode-ai/Wegent) into .claude/skills/develop-wework-plugin in your project. Claude Code loads it when a task matches its description.
How do I install Develop Wework Plugin in Codex?
Run `npx skills add wecode-ai/Wegent --skill develop-wework-plugin -a codex`. Or copy the skill folder (wework/dsh/plugin-developer/codex-plugin/skills/develop-wework-plugin in wecode-ai/Wegent) into .agents/skills/develop-wework-plugin in your project. Codex loads it when a task matches its description.
Can I use Develop Wework Plugin in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add wecode-ai/Wegent --skill develop-wework-plugin -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/develop-wework-plugin, .gemini/skills/develop-wework-plugin, .github/skills/develop-wework-plugin and .opencode/skills/develop-wework-plugin in your project.
What does Develop Wework Plugin need to run?
Going by SKILL.md and its folder, Develop Wework Plugin needs JavaScript for the scripts in its folder. Our summary lists: Node.js.
Does Develop Wework Plugin access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Develop Wework Plugin safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Develop Wework Plugin use?
Develop Wework Plugin is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Develop Wework Plugin use?
About 3.5k tokens (SKILL.md is roughly 14k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 5.7k tokens, read only when the agent opens those files.
What are the alternatives to Develop Wework Plugin?
Skills that share tags, products or a category with Develop Wework Plugin: Claude Code Skill Developer Guide (diet103/claude-code-infrastructure-showcase, 10k stars), Claude Code Command Development (anthropics/claude-plugins-official, 37k stars), Claude Code Plugin Structure (anthropics/claude-plugins-official, 37k stars) and Mistral Vibe Plugin Creator (mistralai/mistral-vibe, 5.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Develop Wework Plugin?
wecode-ai (a GitHub organization) maintains it in wecode-ai/Wegent, which has 868 GitHub stars. The repository holds 8 skills in this directory. The repository was last updated on October 5, 2026.
Source: wecode-ai/Wegent on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.