Pi coding agent extension packages — tools, skills, and integrations.
This repository also configures its own agents. See what pi-extensions tells them →
Skill Claude CodeCodex
A2A (Agent2Agent) Protocol v1.0 bidirectional communication. Distribute tasks to remote agents (Hermes, ADK, LangChain, CrewAI, any A2A peer) and be called by them. Use when delegating work across agents/machines/frameworks, doing parallel multi-agent fan-out, or when another agent needs to call Pi. Trigger on "a2a"…
Skill Claude Code
Delegate bulk implementation, scaffolding, repetitive refactors, and exhaustive test generation to the Antigravity CLI (agy).
Skill Claude CodeCodex
Trajectory-based self-learning loop for Pi. Automatically captures tool-call trajectories, reflects on them to extract transferable learnings (strategy/recovery/optimization), persists to Munin or local JSONL, and injects recent learnings into future sessions. Use evolvereflect after fixing a bug or recovering from an…
Skill Claude CodeCodex
Design KiCad schematics and PCB layouts via the Konnect bridge. Use when the user wants to create/edit KiCad schematics or PCBs, place and wire components, route traces, run ERC/DRC, export Gerbers/BOM/PDF/3D, search JLCPCB parts, use reference circuits, or do manufacturing prep. Trigger on "KiCad", "schematic"…
Skill Claude CodeCodex
Use Munin long-term memory before non-trivial repo work, bug fixes with recurring symptoms, architecture/dependency/setup changes, and when storing durable verified knowledge for future sessions. Search, retrieve, verify, list, store, or share project knowledge across sessions. Use muninsearch for pre-work context…
Skill Claude Code
Google NotebookLM — notebooks, sources, grounded chat, research, and Studio artifacts (Audio Overviews, video, slides, reports, quizzes, etc.) via the notebooklm-py CLI bridge (v0.7.3).
Skill Claude CodeCodex
Enforce using the Obsidian tool (not bash) for all vault file operations. Trigger whenever the user mentions Obsidian, vault, notes, moving files in the vault, archiving notes, or any file operation inside an Obsidian vault path. Also trigger when the agent is about to use bash mv/cp/rm/rename on paths containing…
Skill Claude CodeCodex
Whole-repo audit for over-engineering. Like ponytail-review, but scans the entire codebase instead of a diff: a ranked list of what to delete, simplify, or replace with stdlib/native equivalents. Use when the user says "audit this codebase", "audit for over-engineering", "what can I delete from this repo", "find…
Skill Claude CodeCodex
Harvest every ponytail: comment in the codebase into a debt ledger, so the deliberate shortcuts and deferrals ponytail leaves behind get tracked instead of rotting into "later means never". Use when the user says "ponytail debt", "/skill:ponytail-debt", "what did ponytail defer", "list the shortcuts", "ponytail…
Skill Claude Code
Show ponytail's measured impact as a compact scoreboard: less code, less cost, more speed, from the benchmark medians. One-shot display, not a persistent mode, and not a per-repo number. Trigger: /skill:ponytail-gain, "ponytail gain", "what does ponytail save", "show ponytail impact", "ponytail scoreboard".
Skill Claude Code
Quick-reference card for all ponytail modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /skill:ponytail-help, "ponytail help", "what ponytail commands", "how do I use ponytail".
Skill Claude CodeCodex
Code review focused exclusively on over-engineering. Finds what to delete: reinvented standard library, unneeded dependencies, speculative abstractions, dead flexibility. One line per finding: location, what to cut, what replaces it. Use when the user says "review for over-engineering", "what can we delete", "is this…
Skill Claude Code
Forces the laziest solution that actually works, simplest, shortest, most minimal. Channels a senior dev who has seen everything: question whether the task needs to exist at all (YAGNI), reach for the standard library before custom code, native platform features before dependencies, one line before fifty. Supports…
Skill Claude Code
Anti-slop UI/UX design discipline for AI-generated interfaces. Enforces industrial-design principles (Dieter Rams: honest, thorough to the last detail, as little design as possible) so output is a defensible system, not statistical-default slop (purple glow, shadow-as-texture, missing states) — and equally not the…
Skill Claude Code
Reference design-system presets for ux-design. Anti-slop shortcuts: instead of inventing tokens, elevation, and state contracts from scratch, reuse one of these battle-tested systems (shadcn/ui, Material 3, Radix), drop in a medium-tuned DESIGN.md starter (Web or Mobile), or adopt a style-direction starter (Editorial…
Skill Claude CodeCodex
Web search, content extraction, site crawling, page capture, real-browser interaction (trusted click/type/evaluate in headless Chrome), Gemini web-tier research, image generation, and one-off ChatGPT-web/gateway chat via the pi-web extension. Use when the user needs current web search results, documentation lookup…
Skill Claude CodeCodex
Provides Windows-native tool manipulation: shell profiles, path conversion, command execution, WSL bridge, safety policy, and developer tool discovery. Enables Pi to work correctly with PowerShell, cmd, Git Bash, and WSL.
Skill Claude Code
Capture playbook for the pi-ux render-and-inspect loop: screenshotting a UI you just built via local headless Chrome vs webscreenshot, daemon reachability (localhost vs LAN IP vs host.docker.internal), SSRF-protected daemons, and cloudflared tunnels. Load when setting up render inspection or when a UI screenshot…
Skill Claude CodeCodex
Use when the user wants scheduled, recurring, or periodic tasks in Pi — cron jobs, reminders, jobs at a specific time ("every morning at 9", "weekly monday", "in 30 minutes"), or automation that should run without being asked each time.
Skill Claude CodeCodex
Use when improving pi skills with skillmanage (list/read/patch/create/write/delete/restore). Patch gaps same-turn with minimal diffs; create only recurring procedures.
Skill Claude Code
Reference table for dividing UX work across models in the Pi ecosystem — who defines the DESIGN.md scaffold, who generates variants, who renders and inspects, what reviews. Consult when delegating design work to other models (agy/Gemini, Claude, DeepSeek, Kimi) or when deciding whether the main model should do a step…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: