Instructions for markpasternak/canvas-drop, covering agents.md — canvas-drop, how we work (the loop), autonomous full-scope rounds (mark's preferred workflow), compounding learnings across every agent and project rules (load-bearing — from buildbrief.md).
Instructions for markpasternak/canvas-drop, a project described as: Self-hostable, open-source platform to deploy & share small static web apps ("canvases") inside your org — paste/drag/PUT a folder, get an instant secure URL. Five zero-config backend primitives (KV, files, AI, identity, realtime), an MCP server for…
If you are an agent deploying a canvas, start here. canvas-drop serves a single plain-text contract at {base}/llms.txt — no markup chrome, designed to be dropped straight into context. It is public (served from the docs band, readable without a session) so you can learn the API before you hold credentials.
Install this skill so your coding agent deploys and extends canvases against this instance, first try, without manual correction. It ships in the standard skill format (a SKILL.md with name/description frontmatter and a when-to-use trigger), so an agent loads it automatically when a task matches.
Deploy and extend small web artifacts ("canvases") on a canvas-drop instance. Use when the user wants to ship static HTML/JS to a shared URL, or give a canvas backend capability (key-value, files, identity, AI, realtime) via the zero-config browser SDK. Connect over MCP for an identity-scoped, multi-tool surface, or…