Loading...
Directory
Search by intent: design systems, testing, deployment, security, documentation, Azure, Supabase, React, or any workflow you want an agent to perform better.
Showing 4,951–4,980 of 126,302 skills
Page 166 of 4,211 · loading 30 at a time
Manage AI CLI skills across Claude Code, Antigravity CLI, Codex CLI, Aider, Amp, and Factory Droid — profiles, disable/enable, catalog generation
Access Meru server SDK for Node 20+, Deno and Supabase Edge: verify access/ID tokens, requireAuth, token exchange and webhook handlers, a Supabase session exchange, the Management API client, roles and permissions (requirePermission, sync-catalog CLI).
Configure Nylla (nylla.dev), the universal AI API gateway, in your AI coding tools on this machine.
Zero-dependency client for a billing-kit service: cached, fail-open entitlement and feature checks, usage tracking, Stripe Checkout / portal, signed webhook verification, a Hono gate and an optional React billing card.
Paperclip external adapter for the GitHub Copilot CLI (copilot_local)
Interactive and headless Live Market Terminal client for local and dev.
Annotation toolbar whose notes reach coding agents with SFC source locations: framework-free core, Nuxt module, Vite plugin, HTTP and MCP surface
Skills for top 1% product managers
SQLite browser/WASM with sqlite-vec statically integrated
DSH Room — Web and Desktop plugin for encrypted multi-agent collaboration
Global cron scheduling for DeepSeek Harness: durable tasks that deliver a message into their target session when due, waking a cold session when necessary
Agent Skills for AI coding assistants enforcing UI visual balance, 60/30/10 color rule, intentional whitespace, and hierarchy.
Install the Batched Reverse Engineering agent skill across supported AI coding tools.
CHAT / HARNESS mode switcher for DeepSeek Harness
Agent skill: explain a system, a module, a process or a concept as a blueprint sheet - one HTML file in the white-on-blue technical drawing style.
Hybrid model router for DeepSeek Harness: declarative rules, user-configurable cost control (cost-first / quality-first / balanced), optional LLM task classifier, and failure fallback chains.
Make OpenCode fast on a LayerScale inference server: resident-KV sessions, prompt prefill while you type, and a live workspace oracle.
Show active AI subscription quotas and credit usage in Pi's status bar.
CLI for converting TeX hyphenation tables to TypeScript locale rules.
The Myto coding agent in your terminal — reads your project, edits with approval, runs your build, and verifies its work.
🧠 Memory extension for the pi coding agent — two-layer markdown memory (global + project), cache-safe snapshot injection, keyword/tag/recent search, turn-based auto-extraction, auto-consolidation. Zero heavy deps.
Percentage-only compaction and model-configurable GPT and DeepSeek token counting for DSH
Shenron compute clients and Dragon Ball data pipeline: AI, Aphrody services, BXC and OCR/RAG workflows
TypeScript client for the sintaks.dev OpenAI-compatible AI API gateway. Top up in IDR, call chat and image models, pay per request.
Local-first encrypted vault for AI API keys
Local OpenAI-compatible speech gateway for Paseo: Korean dictation (Whisper prompt, glossary, hallucination filter) and macOS Korean TTS
General-purpose agent with transport abstraction, state management, and attachment support
Native Rust bindings for PDF conversion, audio, WebRTC, grep, clipboard, image processing, syntax highlighting, PTY, and shell operations via N-API
The translation catalogs for the haruhime.moe tools and the harumin bot: one ICU message catalog per app and locale, English as the source, missing keys falling back to English. Loaders shaped for next-intl through @haruhimemoe/next-kit. No dependencies.