Crewship

Crewship is a hosted platform for deploying and running AI agent workflows without managing infrastructure. Developers install its CLI, run a single deploy command against a CrewAI or LangGraph project, and get a live, versioned deployment with real time streaming of each step their agents take, delivered over server sent events. It is aimed at developers moving agent prototypes into production. Every deployment is versioned with instant rollback and separate staging and production environments. Execution traces record each agent action, LLM call and tool use for debugging, and every file an agent produces is collected as an artifact and retrievable through the API. Incoming and outgoing webhooks, signed with HMAC, trigger runs and report completion. Data tables give crews structured storage they can read and write, threads support multiturn conversations with persistent state, and any deployed crew can be connected to Slack. Runs execute in isolated environments with encrypted secrets and scale automatically. Support for LangGraph.js is being added, with Agno, AutoGen, Pydantic AI and OpenAI Agents listed as coming soon. Pricing is published. Free costs $0 per month with one active deployment, 10 versions and 30 runs a month, a six minute maximum runtime and seven days of log retention. Pro is $25 per month with 10 deployments included and $2.50 per month for each extra, unlimited versions, 500 runs a month, 10 concurrent runs, a three hour maximum runtime, 100MB artifacts and priority support. A Team plan with collaboration, role based access and audit logs is on a waitlist, and Enterprise is custom with dedicated infrastructure and an SLA.