@meistrari/tela-skills
v1.1.0
Published
Tela API skills for Claude Code
Readme
@meistrari/tela-skills
Tela API skills for Claude Code. Enables Claude to interact with Tela canvas/prompts directly.
Installation
bunx @meistrari/tela-skillsThe installer will:
- Copy skill files to
~/.claude/skills/tela/ - Optionally set up your Tela API key
Usage
Once installed, Claude Code can use the Tela skills via bun preload:
bun --preload ~/.claude/skills/tela/preload.ts -e "console.log(await tela.listProjects())"Available Functions
Projects
tela.listProjects()- List projects in workspacetela.createProject({ title })- Create a new project
Canvas (Prompts)
tela.listPrompts(options?)- List canvas/promptstela.getPrompt(promptId)- Get a prompt by IDtela.createCanvas(payload)- Create a new canvastela.getCanvas(canvasId)- Get canvas + latest versiontela.updateCanvas(canvasId, payload)- Update canvas (merges config by default)
Helpers
tela.createStructuredOutput(properties)- Create structured output schematela.createVariable(name, type, options?)- Create a variable definitiontela.createMessage(role, content, index)- Create a message with proper formattingtela.markdownToHtml(markdown)- Convert markdown to Tela HTML format
Authentication
The API key is stored at ~/.tela/session. You can set it up during installation or manually:
mkdir -p ~/.tela && echo "YOUR_API_KEY" > ~/.tela/sessionExample: Create Canvas with Structured Output
const canvas = await tela.createCanvas({
title: 'My Canvas',
projectId: 'PROJECT_ID',
workspaceId: 'WORKSPACE_ID',
version: {
title: 'v1',
configuration: {
model: 'gpt-5',
type: 'chat',
temperature: 0,
structuredOutput: tela.createStructuredOutput({
summary: { type: 'string', description: 'Brief summary' },
keyPoints: { type: 'array', items: { type: 'string' } }
})
},
messages: [
tela.createMessage('system', '## Task\n\nAnalyze the document.', 0)
],
variables: [
tela.createVariable('document', 'file', { allowMultimodal: true })
]
}
})License
MIT
