Authentication and authorization is something we are thinking about a lot at the moment, especially for agents that are MCP servers.

Our thoughts here are to handle auth the same way that the MCP spec outlines auth (https://modelcontextprotocol.io/specification/2025-03-26). The key thing is to send authorization requests back to the user in a structured way. For example, if Agent A invokes Agent B, which requires user approval for executing a tool call, that authorization request needs to be piped back to the client, and then propagated back to the agent.

This is technically possible to do with the MCP protocol as it exists today, but I think we will want to add that support in mcp-agent itself so it is easy to pause an agent workflow waiting for authentication/authorization.

One nice property of representing agents as MCP servers is that Agent discovery is the same as server discovery.

This is exactly what we're working on at the moment! (If you're curious about following along progress, check out feature/distributed_tracing branch -- https://github.com/lastmile-ai/mcp-agent/tree/feature/distri...)

The nice thing about representing agents as MCP servers is we can leverage distributed tracing via OTEL to log multi-agent chains. Within the agent application, mcp-agent tracing follows the LLM semantic conventions from OpenTelemetry (https://opentelemetry.io/docs/specs/semconv/gen-ai/). For any MCP server that the agent uses, we propagate the trace context along.

I think that's a fair point. How I envision this to realistically evolve is that MCP servers will expose workflows that handle common tasks. These workflows will be "agentic" because they'll involve LLMs interacting with tools and data, and it will be facilitated over MCP. For example, it would be great to have a "triage" workflow agent exposed by Linear, which in turn might use some MCP servers to make tool calls etc.

I don't know of a usecase where there are such deep recursive agent chains that it becomes unmanageable.

I almost think of mcp-agents as a modern form of scripting – we have agent workflows (e.g. generating a summary of new GitHub issues and posting on Slack), and exposing them as MCP servers has enabled us to use them in our favorite MCP clients.

Agents should be given a time budget, which they can allot to other agents as they see fit. And it's easy to enforce: you kill the process after the allotted time.

Yeah, that was my first thought. I actually wrote a blog post a few weeks ago modeling the point at which agent recursion really gets out of control. https://www.neuralwatt.com/blog/agent-bedlam-a-future-of-end...

That's a valid concern. In `hacker-news-agents`, we're exploring ways to manage the complexity of multi-agent interactions. Perhaps a structured approach to agent communication and state could mitigate some of that 'microservice hell' feel.

This diagram might be helpful: https://pbs.twimg.com/media/GrgGy2jWQAAittE?format=jpg&name=...

That's great to hear! We'd love to know more about your experience and any thoughts you have on it.

Would love your feedback on the Temporal support and the MCP agent server concept which we merged in yesterday