Sealos Canvas logo

Sealos Canvas

Organization
labring
sealos-canvas

Run a local read-only HTML topology UI for a project already deployed by Sealos Skills and return a localhost URL. Use when the user asks to view, inspect, visualize, render, open, or run a local canvas for deployed Sealos resources, mentions ".sealos", Sealos deployment state, Kubernetes resources, topology, resource graph, localhost UI, or invokes "/sealos-canvas".

Overview

Publisherlabring
Repositorysealos-skills
Skill namesealos-canvas
Stars
80
Forks
19
Bundled files
4
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 4 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by labring on GitHub. Read the source before you install it.

Installation

Install the Sealos Canvas AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/labring/sealos-skills.git /tmp/sealos-skills
mkdir -p .claude/skills
cp -r /tmp/sealos-skills/skills/sealos-canvas .claude/skills/sealos-canvas
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Sealos Canvas in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Sealos Canvas on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Sealos Canvas is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Sealos Canvas

Identity and Discovery

  • Owner: sealos-canvas (/sealos-canvas and deployed-resource view, inspect, visualize, topology, or localhost canvas requests).
  • Class: read-only-observation with no downstream mutation handoff.
  • Canaries: CANVAS-DEPLOYED, CANVAS-READONLY, CANVAS-REDACT, and CANVAS-SERVER-LIFETIME.

Scope and Boundaries

Require .sealos/state.json with last_deploy, ~/.sealos/kubeconfig, and live read access. Use only read commands such as kubectl get and kubectl config view; never deploy, update, restart, patch, delete, or apply. Render sanitized resource metadata into .sealos/canvas/index.html and return the exact local URL/cache and server-lifetime contract.

Risk and Confirmation

Keep the read-only boundary, Sealos kubeconfig scope, and Secret/ConfigMap sanitization visible before generator detail. Omit Secret.data and ConfigMap.data from every result. A missing state or kubeconfig stops the request without fallback generation. Never expose Secret data, complete ConfigMap contents, kubeconfig contents, or credentials.

Lifecycle Workflow

For each request, resolve the project, check deployed state and kubeconfig, run the single generator, open the loopback UI, and report sanitized counts. The temporary 127.0.0.1 server runs only for the current process and stops when the process ends or on SIGINT/SIGTERM; stop the server before closing the request.

Progressive Disclosure

Load scripts/generate-canvas.mjs and its owned template/reference paths only after the deployed-state and read-only canaries pass. Do not load or invoke deployment mutation helpers from this entry.

Output, Stop, and Error States

  • success: local_url, html_path, app URL, node count, edge count, sanitized HTML/model, and server-lifetime evidence.
  • stopped: not_deployed, kubeconfig_missing, kubectl_missing, or unavailable read access with the exact safe next action, server_lifetime.status: not_started, and no HTML/server fallback.
  • error: generation or server-start path, sanitized diagnostic category, server_lifetime, and recovery action with secrets and kubeconfig redacted.

Success with --no-serve returns local_url: null and server_lifetime.status: not_started; normal success returns local_url and server_lifetime.status: running with shutdown process exit or SIGINT/SIGTERM.

Handoffs

Canvas consumes the typed direct-result tuple with target: none, inputArtifact: .sealos/state.json last_deploy plus sanitized live summaries, allowedAction: serve a temporary read-only loopback view, failureReturn: sanitized read-only generation/server diagnostic, and responseOwner: sealos-canvas. A deploy response may include the returned link without granting Canvas mutation authority.

Verification

Use generate-canvas.mjs --no-serve, the Canvas contract tests, the canvas evals, and baseline cases canvas-positive-read-only-local-url and canvas-violating-missing-state-or-mutation. Verify 127.0.0.1, local_url, html_path, server_lifetime, read-only commands, sanitized output, and explicit "Stop the server" behavior.

Overview

Render the current repository's deployed Sealos resources as a locally hosted HTML canvas. This skill is view-only: it reads .sealos/state.json and Kubernetes resources, starts a temporary 127.0.0.1 UI server, and returns the local URL to the user.

Hard Rules

  1. Do not deploy, update, restart, patch, delete, or apply resources.
  2. Only use read commands such as kubectl get and kubectl config view.
  3. Use the Sealos kubeconfig at ~/.sealos/kubeconfig.
  4. Do not display Secret data or full ConfigMap contents.
  5. If the project has no .sealos/state.json with last_deploy, stop and tell the user to deploy first with /sealos-deploy.
  6. If kubeconfig or live resource access is unavailable, report the script message and stop.

Workflow

1. Resolve the project

Use the current working directory unless the user provides a local path:

bash
WORK_DIR="$(pwd)"

Confirm this is the intended repository before generating output.

2. Start the local canvas UI

Run:

bash
node "<SKILL_DIR>/scripts/generate-canvas.mjs" --work-dir "$WORK_DIR"

Keep this process running while the user is viewing the canvas. The script writes JSON to stdout after the local server starts.

If ok is false, show the message to the user and end the flow. Do not run fallback discovery, do not deploy, and do not create any other artifact.

If ok is true, use local_url as the primary output.

3. Open the local URL

Open the returned URL with the browser:

text
http://127.0.0.1:<port>/index.html

Then summarize:

  • local UI URL
  • app URL
  • node count
  • edge count

Stop the server process when the user is done viewing the page or when the current task ends.

Output Contract

Success:

json
{
  "ok": true,
  "local_url": "http://127.0.0.1:63220/index.html",
  "html_path": "/abs/path/.sealos/canvas/index.html",
  "node_count": 5,
  "edge_count": 4,
  "app_url": "https://example.sealos.run"
}

Stop condition:

json
{
  "ok": false,
  "reason": "not_deployed",
  "message": "This project has not been deployed by Sealos Skills yet. Run /sealos-deploy first, then run /sealos-canvas again."
}

Visual Target

The locally hosted UI should feel like a topology canvas, not a table:

  1. Top bar with app name, namespace, deployed app URL, generated time, and local UI status.
  2. Dark dotted-grid canvas with deterministic resource-card layout.
  3. Resource cards for app, ingress, services, pods, config, secrets, and volumes.
  4. Dashed or solid SVG connector lines between related resources.
  5. PVC/volume references attached as strips on the related card when possible.
  6. Detail panel for the selected resource.
  7. Events panel with recent related Kubernetes events.
  8. Status colors for ready, sleeping, warning, and failed states.
  9. Lightweight pan, zoom, fit, and reset controls.

Theme extraction is best effort. Reuse the user's repo accent color, font, and radius when easy to detect, but preserve operational readability.

Script

scripts/generate-canvas.mjs is the deterministic entrypoint. It:

  1. Reads .sealos/state.json.
  2. Verifies ~/.sealos/kubeconfig and kubectl.
  3. Reads live namespace resources with kubectl get.
  4. Builds a sanitized canvasModel with app, nodes, edges, events, and theme.
  5. Renders assets/canvas-template.html into an internal .sealos/canvas/index.html cache.
  6. Starts a temporary local HTTP server and prints local_url.

Use --no-serve only for tests or CI checks that should generate HTML without keeping a server process alive.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Sealos Canvas AI skill do?

Run a local read-only HTML topology UI for a project already deployed by Sealos Skills and return a localhost URL. Use when the user asks to view, inspect, visualize, render, open, or run a local canvas for deployed Sealos resources, mentions ".sealos", Sealos deployment state, Kubernetes resources, topology, resource graph, localhost UI, or invokes "/sealos-canvas".

Why use Sealos Canvas on TypingMind?

Because you install it once and use it with any model. Sealos Canvas is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Sealos Canvas in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/labring/sealos-skills/tree/main/skills/sealos-canvas. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Sealos Canvas?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Sealos Canvas?

As many as you like. As long as a model supports skills, you can use Sealos Canvas with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Sealos Canvas AI skill free?

It is published on GitHub by labring. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇