jongcheol-pak/karina-plugins/workspace
Claude Code and agent skill by jongcheol-pak/karina-plugins
- community
Checking that the SKILL.md file is still in the repository.
Operate Karina-managed repos, worktrees, and terminals on Windows through the `karina-cli` repo, worktree, and terminal commands: list and register repos, create, inspect, update, and remove worktrees, start a coding agent in a new worktree, and list, read, wait on, send to, create, split, rename, focus, or close Karina terminals. Use for full handoffs ("hand off", "handover", "give this to another agent", "another worktree", "spawn claude/codex in a worktree"), for reading or waiting on anothe…
Install
npx vskill@latest install jongcheol-pak/karina-plugins/workspaceSKILL.md
The same SKILL.md was published earlier as jongcheol-pak/karina-plugins/workspace.
Karina Workspace (Windows)
This file is a discovery stub, not the usage guide. The full, version-matched reference is served by the karina-cli binary itself — kept out of this file on purpose so it can never drift from the binary that will actually run your commands.
Engage it whenever Karina's repos, worktrees, or terminals matter: registering a repo, creating or removing a worktree, starting a coding agent in a new worktree, handing work off to another agent, reading, waiting on, or sending to another terminal, keeping a worktree's status comment current, and searching what agents said in earlier sessions. It also reads the app's status (status), runs Karina's keyboard-shortcut commands without a keypress (command run), and turns the screen cover on or off per monitor (cover).
Use plain shell tools when Karina state does not matter. This provider is Windows-only.
This surface talks to the running app
The repo, worktree, terminal, command, and cover commands are RPC calls to the Karina app over a named pipe. The app must be running. If it is not, every command answers with a JSON error saying so; the CLI will not launch the app for you.
Resolve the CLI for this session
Choose the executable once and reuse it for every later command:
- If the
KARINA_CLI_COMMANDenvironment variable is set, use its value. Terminals that
Karina opens already set it.
- Otherwise, use the one already on
PATH:karina-clifor the release build,
karina-dev-cli for the dev build (Settings → AI 기능 → Karina CLI puts whichever build you are running there, and each name reaches its own app).
- Otherwise, use the full path of the installed binary.
It is the CLI binary, not the app binary. The app has no console and does not answer these commands; the CLI writes JSON to stdout.
Below, KARINA is a placeholder for the executable you resolved. Substitute it before running anything; do not create a shell variable or run KARINA literally.
If the selected executable cannot run, report its exact error and stop. Do not fall through to another executable, which could silently target a different Karina build.
Load the full guide before running Karina commands
That prints the complete, version-matched guide for the exact binary that will handle your next commands — every command and flag, the handoff rules, the result fields, and the error codes. Read it first, then run the specific command you need.
Don't guess subcommands or flags from memory or from a cached copy of this stub. Prefer --json for agent-driven calls; every response is JSON, including errors.
Drain stdout while the command runs — never wait for exit and read afterwards. The guide is larger than a pipe's default buffer, so a caller that blocks on process exit before reading will deadlock. Use a call that reads and waits together (subprocess.run(..., capture_output=True), Command::output(), execFile).
Orientation commands
These three are read-only and safe to run before you have read the guide:
Why the name starts with karina-
~/.agents/skills/ is a shared directory that several tools install into, so a skill whose name collides with another tool's gets overwritten by whichever was installed last. That is why this skill is named karina-workspace and drives karina-cli repo|worktree|terminal ....
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/workspace/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/jongcheol-pak/karina-plugins/workspace)