major-technology/major
Build and manage Major apps, agents, skills, workflows, and connected resources.
Create and configure a Major agent — a versioned two-file bundle (agent.jsonc + prompt.md) edited on a sandbox — what each field means, how to research connectors and applications before writing the prompt, and the edit/validate/save/publish lifecycle.
Create and edit Major apps — full-stack Next.js apps with frontend, backend API routes, and a live preview — how to create or mount an app sandbox and edit through the workspace tools.
Use this skill when the user asks to debug, investigate, inspect for bugs, fix an issue, or says the app is broken, failing, wrong, blank, errored, or not working. Also use it when you observe a concrete existing failure yourself, including runtime errors, failed tool/API calls, browser console errors, app errors, app logs, actual visual regressions, failing lint/tests/builds, deployment issues, bad data, auth issues, or flaky behavior. Do not use this skill for forward-looking UX polish requests like "make the UX nicer" or "improve the design" unless the user identifies a specific existing broken behavior or regression to verify.
Use this skill when the user is building a brand new app for the first time. Guides connector setup, data layer understanding, and initial planning.
Create and manage Major agent skills — versioned file bundles (SKILL.md + references + scripts) that an attached agent loads on demand — and the sandbox-based edit/validate/save/publish lifecycle.
Use for ANY call against a deployed Major app — listing apps the user can invoke, reading an app's skill (endpoints and request/response shapes), and issuing GET or write requests. Load before querying, calling, or probing a deployed app. Do not use this skill to create, edit, mount, or deploy an app — that is app-builder.
Implements AuthN (identity) operations. Use whenever building anything that requires AuthN
Use for ANY operation against a connected resource or connector — databases, CRMs, email, chat, storage, ads platforms, custom REST/GraphQL APIs, remote MCP servers — and for Major platform resources (managed database, managed file storage, app auth, AI proxy, HTTP proxy). Covers discovering resources, reading their context docs, and the per-connector reference that documents each connector's tools, clients, and API. Load before doing any work that involves a resource.
Creates frontend table views. Use whenever creating any type of table UI on the frontend. Use when doing ANYHTHING that involves tables (using tables to display data).
Explains how webhook access works on deployed Major apps and how to implement webhook endpoints. Use when the user mentions webhooks, needs to receive external HTTP callbacks, or is integrating with services that send webhooks (e.g., Stripe, GitHub, Twilio).
Create and manage Major workflows — JSONC graphs of agent calls, app calls, routers, loops, joins, waits, and human approvals fired by schedules, connector events, or authenticated webhooks — the definition format, node/edge/trigger reference, state expressions, and the sandbox-based edit/validate/save/publish lifecycle.