haojing8312Community·36 Agent Skills Included

WorkClaw

Local desktop runtime for building and running AI employee teams

Runs a local-first desktop workbench where users create, package, install, and chat with AI employee teams without command-line work. Encrypts skills into shareable .skillpack files, manages long-term profile memory, and connects to Feishu for mobile task control. Eliminates cloud dependency, manual configuration, and scattered tooling so non-technical users get working automation faster.

154|1

↗

npx skills add haojing8312/WorkClaw --all -g -y

Available:

Tells the AI agent how to route work through the repo's release, verification, and cleanup skills while enforcing architecture, memory, and skill-packaging boundaries for the WorkClaw runtime.

All Skills in This Repository (36)

Pure Emerald Level Indicators

📦 In Repo

haojing8312

workclaw-release-prep

Analyze git commits and tags to recommend version bumps and draft bilingual release notes.

workclaw-release-readiness

Validate WorkClaw release-sensitive changes and generate a GREEN/YELLOW/RED verdict.

workclaw-change-verification

Map changed files to minimal verification commands and run them.

workclaw-release-publish

Publish a confirmed WorkClaw release with git tag and Windows installer.

workclaw-implementation-strategy

Generate implementation strategies for WorkClaw runtime component changes.

workclaw-cleanup-execution

Execute reviewed repository hygiene cleanup batches and route verification through workclaw-change-verification.

workclaw-repo-hygiene-review

Triage pnpm repo-hygiene findings into categorized cleanup candidates.

using-git-worktrees

Create isolated git worktrees with new branches and run dependency setup and tests.

slack-messaging

Send and read Slack messages via slackcli from the command line.

test-driven-development

Enforce test-first development with failing tests, minimal implementation, and refactoring.

finding-duplicate-functions

Identify semantic duplicate functions in JavaScript/TypeScript codebases via extraction and LLM categorization.

systematic-debugging

Diagnose and isolate root causes of software bugs and test failures.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install WorkClaw?▼

Run `npx skills add haojing8312/WorkClaw --all -g -y` in your terminal to install all skills in this suite globally.

What does WorkClaw actually do?▼

It is a local-first desktop app where you create AI employees, chat with them to execute tasks, and package skills into encrypted .skillpack files for safe distribution.

Can non-technical users run AI employee teams?▼

Yes. WorkClaw uses conversational setup with no command line required, and ships a default team template that is ready to use on first launch.

Which AI models does WorkClaw support?▼

It supports 15+ models from 9 providers, including Claude, GPT, DeepSeek, Qwen, GLM, Kimi, and MiniMax, plus any OpenAI-compatible endpoint.

Is my data private in WorkClaw?▼

Yes. All conversations, employee memory, and skill data stay on your local device with no cloud server, and the Apache 2.0 source code is fully auditable.