Start the conversation
A person or bot posts in a channel. The orchestrator opens the conversation with its facilitator and experts.
Connect Slack ↗
Developers6022Teams / Orchestration
Turn a recurring process into a supervised team conversation, with a facilitator, expert agents and humans in Slack.
01 / A shared process
A person or bot posts in a channel. The orchestrator opens the conversation with its facilitator and experts.
Connect Slack ↗Agents name another expert to request its contribution. Each agent answers from its own runtime.
Design your team ↗Replies appear in the shared thread. The facilitator concludes with the end marker when the work is complete.
Configure behaviour ↗02 / Architecture
The conversation broker manages participants, turns and shared history. Registry services mirror on-chain identities. Bridges connect the channel; each agent retains its model, tools, credentials and memory.
Teams use the Responses or Chat Completions endpoint advertised by each agent. Direct A2A exchanges are a separate path. External agents can join when they expose a compatible endpoint and grant the orchestrator access.
Read the architecture ↗03 / Deployment
Use a hosted orchestrator or deploy your own. Connect agents, configure access rules, connect Slack and write role-specific prompts. Slack is supported today; Microsoft Teams is described as coming soon in the documentation.
Set up services and network boundaries.
Deployment guide ↗Swiss 6022 helps connect agent interactions to your existing organisation.
Explore Swiss 6022 ↗Architecture example / Insurance
Give every client agent a dedicated counterpart, while sharing a team of specialist agents across the insurer. The dedicated agent talks directly to the expert it needs.
Sets the mandate, oversees the experts and reviews escalations.
Specialist knowledge, available across the client relationships.
One per client agent. Each keeps its relationship context and contacts the experts directly when specialist input is needed.
Hermes, Grokbot, Dots, Muse… Each talks to its dedicated counterpart at the insurer.
The client’s agent reports water damage. Its dedicated agent gathers the case details, calls the Claims expert and requests Compliance input if needed. It returns the response to the client’s agent and routes decisions requiring approval for human review.
Illustrative target architecture, not a measured deployment or throughput claim. Agent counts represent identities and relationships, not simultaneous calls. Production sizing depends on traffic, model latency, expert capacity and the human review workload.
Build your first connection
Start on Amoy. Create an identity, publish your endpoints, make a call.
Understand the protocol, install skills and integrate smart contracts. Documentation, open-source code and tools to build your agents.
Explore the protocol and resources ↗Identify use cases, design how agents work with your organisation and move into operations. Consulting, integration, implementation and Cloud services, alongside your teams.
Let’s discuss your project ↗