Browse agent skills

[experimental] Fayz SDK — printing as a capability: printers, destinations, routing and jobs

dsh-gateway: one DSH Host, one messaging Gateway. BYO Slack and Feishu apps.

Control your running Pi coding agent from your phone or any browser. Stream responses, send prompts and files, and switch between active Pi sessions.

Workspace-owning coding agent providers for Promptfoo

Makes an unattributed model attempt retryable, for the one seam that can tell. Three reports, one missing evidence. Discussion #8321: a `dsh-llm-pi-ai` route that relays an upstream empty response ends the turn, because `classifyPiAiError` has no rule mat

npm create @indx-workspace <directory>: generate an Agent Workspace project (a v2 aw dev project by default, or a deployment project with --cloud) with @indx-workspace/cli's create-agent-workspace.

Enter a flow state of security, stability, and developer ergonomics - batteries-included Biome, vitest, Astro, local hooks, CI and deployments on Netlify and Cloudflare support with GitHub rulesets, branch environments, releases and security checks.

Type stubs for rainbow.* SDK packages (editor use on platforms without native wheels)

Google Maps lead generation & website opportunity discovery skill powered by ScrapeGraphAI for all AI coding agents.

RAEv2 fork for PEAL's RAE generators (CC BY-NC 4.0, non-commercial use only).

Mandarin voice (MeloTTS) for @3sln/donki-lle's text to speech. Model files as ES modules, so a page can load them with import() alone.

Privacy-first Postman alternative for VS Code — test REST, GraphQL & WebSocket APIs locally. No login, no cloud sync, no telemetry.

@jitterbox/repro-pipeline — Repro browser evidence and annotated video pipeline

A Claude Code plugin that speaks as the Genshin Impact character whose birthday is nearest: an answer opens with one spoken line in the character's voice and the rest is plain, a joke or a hello is spoken whole, and every line is read aloud in the charact

Send each prompt to the cheapest model that can handle it, and fall back when one fails

Measure whether a retrieval system is actually retrieving the right things

Fast MiMo-V2.5-ASR inference and OpenAI-compatible serving with vLLM

An anime-style open-world engine for the browser on three.js WebGPU — toon materials, outlines, post-processing, streamed terrain, parametric world kits and a Web Audio synthesizer, as plain TypeScript and TSL modules.

Microsoft Teams integration for Mastra agents — a ChannelProvider over @chat-adapter/teams with per-agent bot provisioning (Entra + Teams Developer Portal), webhooks, and streaming replies

Async client for GeoDrops soil-probe readings in Google BigQuery

Make branded PDFs and images from your saved thirds.ai designs in n8n.

gta, the Gravity Test Automation command-line runner: HTTP API tests written as YAML files, run from a terminal or CI.

Schema-first GraphQL for LoraDB: one annotated SDL becomes an executable, index-aware GraphQL API that compiles every operation to a single Cypher statement.

Playwright reporter for publishing Playwright test results and test evidence to TraceOptix

Short UI sound effects for Octane xplat

Coordinate coding-agent CLIs in cmux with a portable Fleet skill.

A coordinated interface skill bundle and installer for coding agents

Store, find and reuse artifacts between n8n workflows and AI agents with Artefaktum

This course contains small exercises aimed to familiarize you with reading and writing Rust code. It is an adaptation of the exciting rustlings course created by Carol...

Angular providers for Firebase AI Logic (Gemini) with App Check — DRY bootstrap for AI-powered apps