← All skill packs

AI Agent Workflow Collection

Skill pack
3 skills0 selected
icesourceSkills in this pack

Pull ALL sub-cards' progress into the CURRENT (parent) conversation in one shot and summarize it for action. Activate on /stray-progress, or phrases like "子卡们怎么样了" / "同步一下子卡进展" / "孩子们干完了吗" / "how are my sub-cards doing". One-shot pull (stray subtasks), never a polling loop.

English

One-shot delegation: spin the CURRENT conversation's small task off into a single stray sub-card (own worktree + session + card), seeded with enough context to run independently — freeing this main session immediately. Activate on /stray-subcard , or phrases like "拆个子卡去修这个,别占 主卡" / "这个小问题开张子卡去搞" / "delegate this to a sub-card". The EXPLICIT invocation with a task IS the user's confirmation — spawn at once, no extra confirm round (unlike stray-subcards, which drafts a multi-item split and waits for approval). Do NOT use the built-in Task tool / background agents as a substitute — those are invisible to the stray dashboard.

English

Fan out parallel SUB-CARDS from the current Claude Code conversation via claude-stray (stray spawn), pull their progress (stray subtasks), and relay a message into one (stray send). Activate when the user says any of: 子卡 / 子任务 / 拆成子卡 / 拆出去并行 / 并行子卡 / fan out / spawn a sub-card / sub-task under stray — or asks how their sub-cards are doing. Also activate to PROPOSE a split when the conversation just produced a list of clearly independent work items (propose only — never spawn without explicit user confirmation). Do NOT use the built-in Task / sub-agent tools as a substitute: those are invisible to the stray dashboard and are NOT sub-cards.

English

Includes Stray Progress, Stray Subcard, and Stray Subcards.

Install with
Claude CodeCodexCursorGitHub CopilotGemini CLI

Supported by the Atlas install prompt. Skill-specific requirements vary.

Pack stats

Skills3CategorygeneralStars0

What’s included

Stray Progress

Pull ALL sub-cards' progress into the CURRENT (parent) conversation in one shot and summarize it for action. Activate on /stray-progress, or phrases like "子卡们怎么样了" / "同步一下子卡进展" / "孩子们干完了吗" / "how are my sub-cards doing". One-shot pull (stray subtasks), never a polling loop.

Stray Subcard

One-shot delegation: spin the CURRENT conversation's small task off into a single stray sub-card (own worktree + session + card), seeded with enough context to run independently — freeing this main session immediately. Activate on /stray-subcard , or phrases like "拆个子卡去修这个,别占 主卡" / "这个小问题开张子卡去搞" / "delegate this to a sub-card". The EXPLICIT invocation with a task IS the user's confirmation — spawn at once, no extra confirm round (unlike stray-subcards, which drafts a multi-item split and waits for approval). Do NOT use the built-in Task tool / background agents as a substitute — those are invisible to the stray dashboard.

Stray Subcards

Fan out parallel SUB-CARDS from the current Claude Code conversation via claude-stray (stray spawn), pull their progress (stray subtasks), and relay a message into one (stray send). Activate when the user says any of: 子卡 / 子任务 / 拆成子卡 / 拆出去并行 / 并行子卡 / fan out / spawn a sub-card / sub-task under stray — or asks how their sub-cards are doing. Also activate to PROPOSE a split when the conversation just produced a list of clearly independent work items (propose only — never spawn without explicit user confirmation). Do NOT use the built-in Task / sub-agent tools as a substitute: those are invisible to the stray dashboard and are NOT sub-cards.

Quick start

  1. Choose your skills. Select individual skills above, or choose the entire pack.
  2. Copy the install prompt. Use Copy install prompt above the list.
  3. Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.

Try it with your agent

After installation, start with a request like:

Example request
Use the Stray Progress skill to help with my task.
Ask me for the context you need before getting started.

Requirements & license

  • A coding agent that can access your project files and install skills.
  • Your agent checks each skill’s instructions, dependencies, and permissions during setup.
  • Reported license for the skills shown: MIT. The author’s terms apply.