Menu
SkillsMP has collected 19 skills from ArdaYILDIZ-DEV/pi-skills-public. Open a skill to review its source and details.
Copy this prompt into the AI assistant you're using.
Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/ArdaYILDIZ-DEV/pi-skills-public.Showing 19 of 19 collected skills.
Guide interactive learning and conceptual understanding through diagnostic questioning, stepwise scaffolding, parallel worked examples, and active recall. Use when the user seeks intellectual understanding rather than direct task execution, asks to teach,…
Design, audit, and harden persistent LLM system prompts, agent personas, and tool instructions: authority, permissions, workflow, failure handling, and output contracts. Use for 'sistem promptu yaz', 'persona oluştur', 'ajan kurallarını düzelt', 'araç…
Diagnose generic frontend design and make targeted, constraint-preserving repairs. Use for 'de-slop this UI', 'what looks AI-generated?', 'ai sloptan temizle', 'arayüzü denetle', or an anti-slop review during a build. For new art direction or a full interface…
Design and build context-specific web interfaces from briefs, mockups, or existing UI: art direction, composition, typography, tokens, and responsive states. Use for landing pages, dashboards, forms, portfolios, 'arayüz tasarla', 'dashboard yap', 'siteyi…
Draft, translate, or revise natural casual internet English for Reddit posts and comments, forum replies, casual DMs, and social posts. Use for 'Reddit postu yaz', 'yorum/comment yaz', 'casual English', 'bunu doğal İngilizce yaz', 'make this sound like a real…
Draft, translate, or revise natural professional English for work emails, Slack-style messages, pull requests, code reviews, issue reports, technical write-ups, project updates, handoffs, and support communication. Use for 'professional English', 'iş…
Spawn, steer, and coordinate interactive Pi subagents (scout, researcher, worker) in background tmux panes. Use to delegate codebase reconnaissance, web research, or scoped implementation; includes 'subagent kullan', 'alt ajan başlat', 'scout çalıştır',…
Design, build, restyle, or audit Apple-inspired websites, landing pages, dashboards, forms, and web components using semantic colors, clear typography, responsive layouts, and accessible interactions. Use for 'Apple-style web page', 'design interface with…
Browse, search, and read Reddit with rdt-tidy compact tables for posts, comments, subreddits, and users. Use for a Reddit URL, post/comment reading, subreddit or user lookup, or Reddit search; includes 'reddit'e bak' and 'redditte ara'. Not for publishing,…
Audit, prune, and standardize comments and API docstrings across TypeScript, JavaScript, Python, Go, C, and shell to match idiomatic industry standards (TSDoc, JSDoc, Google-style, Godoc, Doxygen) with zero code logic mutation. Use when the user asks to audit…
Draft, review, split, and safely execute Conventional Commits 1.0.0 messages with correct types, scopes, imperative English subjects, and no AI-attribution trailers. Use to prepare a commit message, inspect git status/diff, split a changeset, or commit…
Build, review, test, and package Pi agent TypeScript extensions using supported ExtensionAPI tools, commands, events, and TUI widgets. Use to add `registerTool`, `registerCommand`, lifecycle handlers, extensions under `.pi/extensions`, or a pi package;…
Conduct reproducible, disk-backed multi-source research with scoped questions, fetched primary sources, exact citations, and a concise report. Use to investigate, fact-check, compare technologies, survey an ecosystem, or find authoritative sources; includes…
Run and manage user-requested long-lived or interactive terminal work in isolated tmux sessions without blocking chat. Use for background servers, REPLs, gdb, log tails, and 'arka planda çalıştır', 'tmux oturumu aç', or 'sunucuyu arkaya at'. Not for a short…
Verify a requested change or claim with fail-first evidence, targeted checks, live output, and exit codes. Use for 'çalışıyor mu', 'testleri çalıştır', 'doğrula', 'kanıtla', and before declaring a bugfix, feature, refactor, or API change complete. Not for…
Draft, audit, and tighten repository instruction files such as AGENTS.md and CLAUDE.md: scoped constraints, verified commands, and non-obvious conventions. Use for 'AGENTS.md yaz', 'CLAUDE.md oluştur', 'repo kurallarını sadeleştir', stale agent instructions,…
Draft, review, and improve task prompts using explicit outcomes, relevant context, boundaries, and acceptance criteria. Use for one-shot requests or bounded multi-turn tasks, including 'prompt yaz', 'promptu iyileştir', and 'bu isteği ajana nasıl anlatırım'.…
Write, audit, restructure, or translate project READMEs in English or Turkish using repository evidence, runnable quickstarts, and audience-appropriate sections. Use for 'README yaz', 'README düzenle', 'beni oku hazırla', or documenting a CLI, library, app,…
Create, audit, and improve agent skills: SKILL.md or Pi flat-file skills, trigger descriptions, workflows, and bundled resources. Use when the user asks to write or revise a skill, diagnose missed or excessive triggering, or package repeatable expertise;…