Build with and use Pi, the minimal terminal coding harness. Use for installing Pi, configuring providers/models/settings, creating Pi skills/extensions/packages/themes/prompt…
After harvester-query returns a hit, render the result as a PI-lawyer-grade brief block (bluebook citation, paste-ready pull quote, NPS class-of-victim/class-of-harm analysis,…
Harness-specific guidance — when to dispatch via Claude Code's Agent tool vs /pi:agent, how to monitor a dispatched specialist, and how to interpret /pi:status.
How to use pi-cc-plugin slash commands (/pi:agent, /pi:status, /pi:result, /pi:cancel, /pi:setup) to delegate tasks from Claude Code to pi-subagents.
统一项目分析 CLI 工具。支持 DAG 调度、LLM 批量任务、依赖图构建、测试分析/修复、文档生成、代码审计、Web Dashboard。与 project-index 功能完全一致，推荐作为统一入口。
Use when embedding `@mariozechner/pi-coding-agent` as a coding-agent runtime in Node scripts, writing pi extensions (plugins) with `pi.registerTool` / `pi.registerCommand` /…
Reference for customizing and configuring pi-coding-agent (pi). Covers settings files (settings.json), extensions, skills, prompt templates, package management, and programmatic…
Calculate and display the token usage, turns, and estimated cost for the active session and subagent runs by parsing the harness session logs.
Register custom Pi Agent model slugs so saved OpenRouter variants resolve correctly.
Dynamic Context Pruning extension for pi. Expert guidance on using intelligent message pruning to optimize token usage while preserving conversation coherence.
Look up Pi coding-agent documentation in the local Miyo-indexed folder labeled `pi`. Use when answering questions about the Pi terminal coding harness, the `pi` CLI, extensions,…
Build, debug, and package pi coding-agent extensions. Use when the user wants to create or modify a pi extension, register custom tools or commands, handle lifecycle events, or…
Persistent Codex-style goal tracking for pi. Use when the user explicitly asks to set, continue, audit, pause, resume, complete, or inspect a long-running goal.
Raspberry Pi health monitor. Check CPU temperature, throttling status, voltage levels, memory/disk usage, fan RPM, overclock detection, and power issues.
When encountering pi tool limitations, propose focused extensions that solve one problem each.
Searches for PIDO and PIASK comments in the codebase, answers questions and executes requested changes, then removes resolved markers.
Drafts a structured personal injury client intake form for initial consultations covering client identification, incident details, injury/treatment history, insurance, prior legal…
Use when interacting with the user by voice through @p8n.ai/pi-listens, Sarvam AI speech-to-text, or Sarvam AI text-to-speech.
Konfiguruje modele AI w pi poprzez models.json. Domyślnie edytuje plik w repo (dotfiles/pi/models.json), nie w ~/.pi/agent.
Create agent monitors and lifecycle hooks for pi. Monitors watch agent output via a cheap model and inject reminders when milestones are reached.
Orchestrate multiple AI models (GLM, MiniMax, etc.) as workers using Pi Coding Agent with Claude as coordinator.
Analyzes existing plugins to extract their capabilities, then adapts and applies those skills to the current task.
Authors system prompt additions for pi coding agent. Use when extending pi's default prompt with custom principles, guidelines, or project-specific rules via SYSTEM.md or…
Use when iteratively refining a design doc, plan, or specification through repeated pi reviews until quality converges
How the main Claude Code thread should treat the output returned by the pi rescue subagent and /pi:result
Use when needing external code review of git changes or design document review before implementation, or when the user asks for a pi review
Send a message to another running pi session via its Unix socket. Use when the user wants to hand off context to another session, continue a conversation elsewhere, or notify…
将具体执行委派给 pi coding agent + ZenMux inclusionai/ling-3.0-flash（廉价执行模型），Claude 负责任务匹配判断、任务书编写、驱动（一次性 pi -p 或 tmux 交互长程）、硬超时重试、独立验收与失败裁决。含 ling-3.0-flash 专属最佳实践（3D…
Delegate work to builtin or custom subagents with single-agent, chain, parallel, async, forked-context, and intercom-coordinated workflows.
Use when work should be tracked in persistent shared pi-tasks lists: multi-step plans, resumable work, handoffs, parallel agents, task claiming, task status updates, or when the…
Give Pi Agents a safe web-search and fetch workflow using the installed pi-web-access package.
Datenschutz-Folgenabschätzung (DSFA) nach Art. 35 DSGVO erstellen: Schwellwertanalyse, BfDI-Blacklist-/Whitelist-Abgleich, strukturierte DSFA im eigenen Hausformat.
Gera um Relatório de Impacto à Proteção de Dados Pessoais (RIPD — Art. 38 LGPD, equivalente brasileiro à DPIA) em formato de casa para nova funcionalidade, produto ou atividade de…
Pianifica una successione ereditaria con calcolo quote legittime, imposte di successione, franchigie e adempimenti.
Piano implementativo — spezza un lavoro complesso in task bite-sized con percorsi esatti, comandi e verifiche. Ogni task 2-5 minuti.
Re-anchor the discipline that a tool, library, framework, language, or approach is chosen to fit the actual problem — not reached for out of habit, availability, incumbency, or…
Use when choosing which GTM tools to use for a task, play, or client engagement — "what tools should I use for X", "build a stack for this play", "which tool does Y", or when…
GitHub 원격 저장소에서 사용자에게 assign된 이슈 목록을 가져와 작업할 이슈를 선택하게 한 뒤, 선택된 이슈의 내용을 프로젝트 메모리에 저장하는 스킬. 사용자가 '이슈 선택', '작업할 이슈', '이슈 골라줘', '내 이슈 보여줘', '이슈 목록', 'pick issue', 'my issues',…
Use when starting a new task, feature, or project where the user has domain knowledge the agent lacks, before planning sessions, or when requirements feel ambiguous and the agent…
Asks a few plain questions about the work, then names the one desktop app to run local AI models in, from LM Studio, Goose, Open WebUI, AnythingLLM, OpenWork or Odysseus.
Pick a Sui Overflow 2026 track (Agentic Web, DeFi, Walrus, DeepBook). Use when the user wants a track recommendation for Sui Overflow.
Fetch and rank open GitHub issues by community engagement, present the top 3 candidates, and plan implementation for the selected issue.
Pick the next shell whose dependencies are satisfied and carry it through planning: expand, refine, self-improve, halt.
Choose professors, mentors, classes, or office-hours targets by human energy, fit, rigor, and availability rather than title worship, GPA spreadsheets, or department mythology.
Pick the right library for a given frontend task from a curated, opinionated list — numbers, OTP inputs, charts, command menus, virtualization, drag and drop, toasts, state,…
Generates multi-themed food collages (Burgers, Pizzas, Sushi, Salads, Desserts) with circle-masked images.
Use when creating or managing picklist fields: choosing between Global Value Sets and object-local picklists, configuring controlling and dependent field relationships, managing…
Picklist governance for Salesforce admins — Restricted vs Unrestricted, Global Value Sets vs local picklists, value deactivation that breaks reports, dependent picklists with…
Use when picklist data quality has degraded: unrestricted picklists accepting garbage API values, dependent picklist relationships bypassed by integrations, record type picklist…
Pickrr integration. Manage data, records, and automate workflows. Use when the user wants to interact with Pickrr data.
captures/ 큐에서 자연어로 항목을 선택하고, 변경 요청 시 /mst:plan --from-picks로 자동 전환합니다.
Retrospective on past token and prediction market picks - what hit, what flopped, what the score is
Claim the next available task from Linear (or local phases.yaml), create a git worktree, and inject relevant learnings.
Picky Assist integration. Manage data, records, and automate workflows. Use when the user wants to interact with Picky Assist data.
Picoclaw runtime traffic monitoring baseline for lightweight AI gateway proxy inspection, egress detection, and posture integration.
Agents Checking Agents — multi-agent orchestration with cross-validation using Claude Code Agent Teams. Fresh agents for each phase — no agent reuse.
Use when the user invokes /pics or demands proof for every claim — forces inline evidence (diff, test output, command output, file:line) for every factual statement in the reply.
Quickly shorten URLs and generate QR codes via PicSee (picsee.io). After logging in, you can also view analytics and history.
Nutze diesen Skill, wenn der User den Batch-Run für seine Gym-Fotos triggert (z. B. "gym batch", "analysiere gym", "verarbeite Trainingsbilder").
Use when a user asks about P&ID drawings, equipment, pipes, instruments, control loops, process flow, safety checks, or wants P&ID context exported for an LLM.