Ask HN: Is anyone building agents with MCP (or it's just for Claude integration)

1 points by resiros 1y ago ↗ HN

MCP is great for integrating with Claude/Cursor, but building production agents with it does not make sense to me: You don't have access to the server's prompts, you lack observability, and can't debug.

Just had a convo with my teammate about this and curious what hn thinks.

1 comment

[ 3.0 ms ] story [ 15.1 ms ] thread

resiros 1y ago ↗

Most of the work when trying to build a reliable agent is 1) determining which tools do you provide to the llm 2) how to describe the tools, their interface. MCP gives you pre-built tools that you can't change neither the interface nor the descriptions/prompts.

There is value in quick integration (or at least the promise of), but I don't see why it would be used when building an agent.