Subagents
- 22 installs
- 5k repo stars
- Updated August 4, 2026
- dtyq/magic
Delegate independent subtasks to parallel subagents (explore for deep search, shell for commands) instead of doing large research or exploration inline.
About
subagents guides an agent on when and how to delegate work to parallel subagents rather than doing everything inline. It suits independent subtasks, large research or exploration that would otherwise clutter the main thread, and cases needing a specialized agent type such as explore for deep search or shell for system commands. A solo builder uses it to keep a long agent run focused and to fan out work that has a clear deliverable and no dependency on the current thread.
- Run independent subtasks in parallel
- Keep large research separate from the main thread
- Use specialized agent types like explore and shell
Subagents by the numbers
- 22 all-time installs (skills.sh)
- Ranked #10,137 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/dtyq/magic --skill subagentsAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 22 |
|---|---|
| repo stars | ★ 5k |
| Last updated | August 4, 2026 |
| Repository | dtyq/magic ↗ |
What it does
Delegate independent subtasks to parallel subagents (explore for deep search, shell for commands) instead of doing large research or exploration inline.
Who is it for?
Fanning out independent subtasks to keep the main thread clean
Skip if: Tightly coupled sequential steps
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10