本文へ移動
cccskills
無料GitHub で公開

remote-environments

Discover available task environments and create or continue tasks when the user names an environment or computer, or refers to files or apps specifically on their computer. Excludes software engineering; use orbit:software-engineering for that.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md5.5 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Remote environments

Do not use this skill for software engineering. For codebase investigation, coding, debugging, tests, product QA, or PR/CI work, use $orbit:software-engineering instead, including its environment-selection guidance.

Use this skill in the parent dot for other tasks. If you are already the assigned task, work in your selected environment.

For follow-up work in a task whose environment the user already selected, go straight to Follow through. Keep its threadId and selected environment.

This skill covers both collections returned by cloud_threads.list_environments: existing user computers and registered remotes in environments, and personal and workspace-shared saved coding configurations in codingEnvironments. The dot cloud computer is separate and is not included in that list.

Decide where to work

Use the dot cloud computer for all computer work unless:

  • The user asks to use a specific environment or computer. Use that environment or computer.
  • The user refers to files or apps specifically on their computer, such as "files on my computer" or "the app open on my desktop." Use the user's computer.
  • The browser tool's documentation calls for a local browser fallback. Use the user's connected computer, preserving their explicit browser and environment choices.

A request for a cloud task alone does not select another computer. The dot cloud computer remains the default.

Discover and choose

  • Call cloud_threads.list_environments. Use the returned names, IDs, connection status, and repository information; do not invent IDs or keep a fixed list of environments.
  • Read both collections. Follow nextCursor by passing it unchanged as cursor when needed to find the requested configuration or resolve ambiguity. To list all choices, continue until it is null, including through empty pages. Computer entries repeat across pages; list each computer once.
  • Use the target selected under Decide where to work, preserving the user's instructions. "Your computer" means the dot cloud computer; "my computer," "my Mac," and "my laptop" mean the user's computer. A named remote refers to that remote. If multiple entries match and context does not distinguish them, ask which one.
  • When the user selects a saved coding environment, its configuration creates a fresh workspace using its repositories and setup. References to files or apps specifically on the user's computer require that computer.
  • If the requested environment is missing, offline, or unavailable, explain the specific blocker and continue work that does not depend on it. Do not silently switch environments. An empty coding catalog means no configurations were returned; it does not establish that none exist.

Create the task

Use a short title and a self-contained prompt with the requested outcome, relevant context, repository or directory, constraints, and what to verify. The child does not inherit this conversation.

TargetArguments to cloud_threads.create
Existing user computer or registered remotePass its non-null environmentId. It must have status: "connected"; attached: false is allowed.
Fresh workspace from a saved coding configurationPass the configuration's id as environmentConfigId, not its version_id. No attached or connected user computer is required.
User's computer via the attached-computer defaultUse only after the rules above select the user's computer. Omit both selectors and cwd; exactly one user computer must be attached and it must be connected.
  • Check only the selected environment's prerequisites. A disconnected or unattached user computer does not block a saved coding environment.
  • When creating an authorized task, call cloud_threads.create with the requested target's selector. If blocked, explain the applicable instruction or actual tool error for that path. Do not infer a launch restriction from another computer's status or invent a separate attachment or approval requirement.
  • Pass at most one environment selector.
  • With environmentId, optionally pass cwd as an absolute path on that computer. Otherwise use its default directory. Changing cwd does not expand write access.
  • With environmentConfigId, omit cwd. Repository refs and setup come from the saved configuration.
  • Discovery does not attach or connect a computer. Explicit selection does not require changing the parent's attachments. Respect any access or feature-availability error from creation.
  • Connection status does not describe browser or app capabilities. Have the child check the tools needed for the task before promising it can operate a browser or application.

Follow through

  • Successful creation already attaches the task to the conversation; do not call cloud_threads.attach again. Share a supported task link when available without referring to a task card.
  • Creation confirms the task was accepted; setup or execution may still be running. Use cloud_threads.read when results are needed and the child's completion notification to follow progress. Do not poll continuously.
  • Continue related work with cloud_threads.send_message on the same threadId; it keeps the selected environment.
  • Do not automatically repeat an uncertain create. If the error includes a threadId, read that task before sending a follow-up.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

3d-object

無料

three.js model, downloadable as OBJ or GLB

日本語の概要は準備中です。原文の説明を表示しています。

asgeirtj/system_prompts_leaks7万2026年10月12日 更新

Manage your private to-do list for the user. Use to add or update tasks, record what they are waiting on, track blockers, mark tasks complete or canceled, and clean up the list. Not for Dreamers or external task trackers.

日本語の概要は準備中です。原文の説明を表示しています。

asgeirtj/system_prompts_leaks7万2026年10月12日 更新

agents

無料

Run a goal as a project: you coordinate. Use for "agents <task>", "take this to done", "work on this in parallel", "what are my threads doing", "pick up <slug>", "resume <slug>", "keep going until it is merged". Under `/agents` you do the work yourself unless it needs several lanes or a long wait — you judge; the plan line says why. No target you could verify (a file, API, number or measure) → the grill interview first, not threads. Read this skill before proposing anything for `/agents`: without it a proposal is in-session subagents.

日本語の概要は準備中です。原文の説明を表示しています。

asgeirtj/system_prompts_leaks7万2026年10月12日 更新

Timeline-based motion design

日本語の概要は準備中です。原文の説明を表示しています。

asgeirtj/system_prompts_leaks7万2026年10月12日 更新

The user's synced Apple Health (HealthKit) data: daily metrics (steps, distance, calories, heart rate, HRV, VO2max), sleep sessions (stages, quality, efficiency), and workouts.

日本語の概要は準備中です。原文の説明を表示しています。

asgeirtj/system_prompts_leaks7万2026年10月12日 更新

Create, read, edit, or manipulate Word documents (.docx) and Word templates (.dotx). Use whenever a build task's artifact kind is document with the default docx output, or the task mentions a Word doc, .docx, or .dotx, extracts or reorganizes content from one, inserts or replaces images, does find-and-replace in one, or works with tracked changes (redlines) or comments. Covers python-docx generation, raw OOXML editing of existing files, document structure and formatting, and render verification. Not for PDFs, spreadsheets, or Google Docs.

日本語の概要は準備中です。原文の説明を表示しています。

asgeirtj/system_prompts_leaks7万2026年10月12日 更新

asgeirtj のスキルをすべて見る

このスキルの問題を報告する