canvas

canvas is a skill for Claude Code from dvdsgl/claude-canvas. It costs 36 tokens per session (776 once invoked), scanned A, original, MIT.

A toolkit for showing and controlling interactive terminal displays, such as calendars, documents, and flight bookings. These displays let an agent present information or support guided selections and edits.

In plain words
What is it for?
Use it to display calendars, choose meeting times, view or edit documents, preview emails, compare flights, and select seats.
Why use it?
It gives terminal-based tasks a visible interactive interface instead of requiring every choice or document view to happen through plain text.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the canvas plugin — 4 skills, 1 command shipped together

Good fit Use it to display calendars, choose meeting times, view or edit documents, preview emails, compare flights, and select seats.

Compare 6 skills from other repositories ↓
About the project

Claude Canvas is a terminal user-interface toolkit that gives Claude Code an interactive display separate from its main session. It is for spawning terminal interfaces for tasks such as managing email, calendars, flights, and bookings.

dvdsgl/claude-canvas · 1,511 stars · on GitHub

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add dvdsgl/claude-canvas
Claude Code
/plugin install canvas

Made for: Claude Code.

Or install canvas, the plugin that ships this one along with the rest of its 4 skills, 1 command.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for canvas

README.md
[![agentmods](https://agentmods.dev/badge/skills/dvdsgl/claude-canvas/canvas/github.svg)](https://agentmods.dev/skills/dvdsgl/claude-canvas/canvas)
Your own site
<a href="https://agentmods.dev/skills/dvdsgl/claude-canvas/canvas"><img src="https://agentmods.dev/badge/skills/dvdsgl/claude-canvas/canvas/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for canvas

Your own site · 80×15
<a href="https://agentmods.dev/skills/dvdsgl/claude-canvas/canvas"><img src="https://agentmods.dev/badge/skills/dvdsgl/claude-canvas/canvas.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 776 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00036 $0.00776
Opus 5 $0.00018 $0.00388
Sonnet 5 $0.00007 $0.00155
Haiku 4.5 $0.00004 $0.00078

Measured 9d ago against content hash e5c40c0e8e47, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

canvas scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

canvas/skills/canvas/SKILL.md · 116 lines

How it starts

The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Canvas TUI Toolkit

Start here when using terminal canvases. This skill covers the overall workflow, canvas types, and IPC communication.

Example Prompts

Try asking Claude things like:

Calendar:

  • "Schedule a meeting with the team next week"
  • "Find a time when Alice and Bob are both free"

Document:

  • "Draft an email to the sales team about the new feature"
  • "Help me edit this document — let me select what to change"

Flight:

  • "Find flights from SFO to Denver next Friday"
  • "Book me a window seat on the morning flight"

Overview

Canvas provides interactive terminal displays (TUIs) that Claude can spawn and control. Each canvas type supports multiple scenarios for different interaction modes.

Available Canvas Types

Canvas Purpose Scenarios
calendar Display calendars, pick meeting times display, meeting-picker
document View/edit markdown documents display, edit, email-preview
flight Flight comparison and seat selection booking

Quick Start

cd ${CLAUDE_PLUGIN_ROOT}

# Run canvas in current terminal
bun run src/cli.ts show calendar

# Spawn canvas in new tmux split
bun run src/cli.ts spawn calendar --scenario meeting-picker --config '{...}'

Spawning Canvases

Always use spawn for interactive scenarios - this opens the canvas in a tmux split pane while keeping the conversation terminal available.

bun run src/cli.ts spawn [kind] --scenario [name] --config '[json]'

Parameters:

  • kind: Canvas type (calendar, document, flight)
  • --scenario: Interaction mode (e.g., display, meeting-picker, edit)
  • --config: JSON configuration for the canvas
  • --id: Optional canvas instance ID for IPC

IPC Communication

Interactive canvases communicate via Unix domain sockets.

Canvas → Controller:

{ type: "ready", scenario }        // Canvas is ready
{ type: "selected", data }         // User made a selection
{ type: "cancelled", reason? }     // User cancelled
{ type: "error", message }         // Error occurred

Read the full file on GitHub · 116 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 116 lines · 36 tokens per session scan A e5c40c0e8e47

Subscribe to this mod's changes

canvas is a skill published in the GitHub repository dvdsgl/claude-canvas (1,511 stars, last pushed 8mo ago), licensed MIT. It adds 36 tokens to every session and 776 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.