Pi Prompting

Internal scaffolding for writing prompts that DeepSeek (and other models Pi might run) can execute reliably. You won't invoke this directly. It's guidance for the pi-rescue skill when it needs to delegate coding, review, or research tasks to Pi. The core idea is to prefer concrete checklists over vague instructions, always define what "done" looks like, and use stable XML-style blocks for task, output contract, grounding rules, and action safety. Ships with prebuilt review prompts and a recipe for splitting complex asks into single-purpose runs. Useful if you're building skills that orchestrate LLM subtasks and want a tested pattern for keeping those prompts tight and grounded.

skills.sh installs: 2.9K
GitHub stars: 100
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add agents365-ai/365-skills --skill pi-prompting --agent claude-code

Installs into .claude/skills of the current project.

Files

Pi Prompting

Use this skill when pi:pi-companion-forwarder needs to ask Pi for help on a non-trivial coding or investigation task.

Pi runs whatever model the user has configured. By default this plugin targets DeepSeek V4 (Pro for review, Flash for everyday tasks). The guidance below assumes the prompt may run on either a non-reasoning model (Flash) or a reasoning model (Pro). Bias toward concrete, numbered checklists rather than abstract instructions — both model classes follow those reliably.

Core rules:

Prefer one clear task per Pi run. Split unrelated asks into separate runs.
Tell Pi what done looks like. Do not assume it will infer the desired end state.
State explicit grounding rules whenever an unsupported guess would degrade quality.
Use stable XML-style tag blocks so the prompt has obvious internal structure.
Keep prose short. Prefer numbered checklists over paragraphs for procedural work.

Default prompt recipe:

SKILL.md on GitHub