Understand the project
Practical overview
When it fits
Use Openhermit when agent operation, durable state, isolated execution and fleet-level administration matter more than a single local agent session.
Who it is for
- Teams operating multiple persistent AI agents
- Developers building self-hosted or cloud-oriented agent services
- Organizations that need centralized fleet configuration and isolated agent workspaces
Documented uses
- Deploy AI agents as managed production services
- Keep sessions, memories, instructions, skills, schedules, secrets and MCP configuration in durable centralized state
- Run agent code and tools inside per-agent sandboxes
- Apply skills, instructions, MCP servers and secrets to individual agents or fleets
- Connect agents to supported communication channels
Before you choose
Requirements & limits
The repository is public, open source and MIT licensed. The documented deployment targets include self-hosted Docker, E2B, Daytona, cloud and server environments. No pricing or account requirement is established here.
- PostgreSQL is used for shared internal state, while workspace files remain in per-agent sandboxes.
- Sandbox choices include self-hosted Docker, E2B and Daytona.
- The gateway supports HTTP, SSE and WebSocket RPC transports.
- Channel adapters include Telegram, Discord and Slack, with additional package-installed adapters listed for Signal, WeChat and WhatsApp.
- Deep Research requires a configured web provider and uses model calls that can consume substantial token budgets.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Evidence and transparency
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 3 October 2026
- Original website title
- HCF-STUDIOS/openhermit · Official homepage title · retained in its source language