Sim

Sim is an open-source AI workspace for teams that build, deploy, and manage AI agents. It is built for creating agentic workflows and for turning plain-English requests into working agents that can automate real work.

The product supports three ways to assemble agents: visually, conversationally, or with code. Its workflow builder lets users wire blocks, models, and integrations into agent logic, including one agent or multiple agents running in parallel. Sim also provides a set of named building blocks and controls, including an agent router, function, condition, API block, workflow, parallel, guardrails, evaluator, and loop. For context handling, it stores data semantically in tables, files, and knowledge bases so agents can read from that information and ground their answers in the user’s own data.

Monitoring is part of the workspace as well. Sim says users can trace each run block by block, view full logs, and see the real cost of what ran. The page also gives an example of an agent workflow that enriches a new lead and posts it to a sales channel, which illustrates the kind of operational automation the product is meant to support.

Sim is delivered as a workspace and also lists self hosting among its product links. It connects to 1,000+ integrations and every major LLM, with named integrations including Slack, HubSpot, Salesforce, Notion, GitHub, Gmail, Jira, Linear, Supabase, and Stripe. The site describes it as SOC2 compliant and production-ready for teams of every size, and it says it is trusted by over 100,000 builders.