Menu
Agent Skill for spawning a subagent via Pi Coding Agent. Use when spawning a subagent with various models.
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/CXwudi/agent-config --skill pi-subagentThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Based on SOC occupation classification
Showing SKILL.md
| name | pi-subagent |
| description | Agent Skill for spawning a subagent via Pi Coding Agent. Use when spawning a subagent with various models. |
pi --help resolves and returns the help menu
Otherwise stop and tell the user to install Pi Coding Agent
Run pi --list-models to see all available models (or pi --list-models <provider> to see models for a specific provider)
deepseek-v4-flash from deepseek provider.--provider to specify the provider.openai, use openai-codex insteadanthropic--thinking max, it will fallback to the highest available thinking effort for the model.pi can read the project wise AGENTS.md file and .agents/skills/pi has a copy of the same user-wise AGENTS.md or CLAUDE.md file, and a copy of same agent skills at ~/.agents/skills/-c option can be used to continue from the previous chat. Use it when you only spawn one subagent and want the same subagent to continue further works.Agent Skill for Word, Excel, PowerPoint, and other office works.
Agent Skill for entering orchestration mode. Use when the user want you to become an orchestrator of multiple subagents.
Agent Skill for syntax-aware code search using `ast-grep`. Use when searching for code. Good at finding code patterns, structures, or relationships that are difficult to express with plain text search