OpenMolt
import OpenMolt from 'openmolt';
const om = new OpenMolt({
llmProviders: {
openai: { apiKey: process.env.OPENMOLT_OPENAI_API_KEY },
},
});
const agent = om.createAgent({
name: 'Comedian',
model: 'openai:gpt-4o-mini',
instructions: 'You are a witty stand-up comedian.',
});
const result = await agent.run('Tell me a joke!');
console.log(result);OpenMolt uses a scope-based permission model. Your API credentials are stored server-side and rendered into HTTP requests via Liquid templates — the LLM only receives the results of tool calls, never your raw API keys or tokens. Grant only the scopes each agent needs.
Every aspect of an agent is configurable: instructions (or load from a file), model provider, model config (temperature, thinking mode, token limits), output schema, max steps, and per-integration scope restrictions. Build exactly the agent you need.
Agents maintain long-term and short-term memory stores. Provide an onUpdate callback to persist memory to a database or file. Agents can update their memory mid-run and carry context across multiple sessions.
Schedule agents to run on an interval (e.g., every 20 minutes) or on a cron-style daily schedule with timezone support. Perfect for daily reports, monitoring, automated content publishing, and more.
Schedule a Gemini agent to pull metrics from Stripe every morning, generate a summary, and post it to your Slack channel — automatically.
Describe your content strategy in natural language. An agent writes the blog post, generates images with DALL-E, and saves everything to disk.
Automatically draft replies to incoming emails based on your guidelines. Review drafts in Gmail before sending — no more starting from scratch.
Triage issues, auto-label PRs, post release notes to Slack, and generate changelogs — all triggered by your existing CI/CD pipeline.
Monitor Shopify orders, update inventory in Airtable, send order confirmations via Twilio, and report daily revenue to a Notion dashboard.
Scrape product pages with browser-use, analyze competitors with web search, compile findings into a structured Notion database.