MCP Doesn't Suck. Your Agent Does. — Jan Čurn, Apify
Oct 2, 2026AI Engineer
Summary
This talk argues that the perceived failures of the MCP protocol are actually due to naive agent harness implementations. It proposes three key fixes: sub-agents for managing complexity, progressive tool discovery to avoid context bloat, and Code Mode for better code generation. The talk also makes a case for using MCP for remote, authenticated access and CLIs for local access, demonstrating this with mcpc, an open-source universal CLI client for MCP that supports stdio and remote servers, OAuth, persistent sessions, grep-based tool discovery, async tasks, and x402. Connector Evals show mcpc outperforming raw MCP and CLIs.
More from the Agent Memory and Harness Engineering trail
- From Context to Memory: Your Agents Need a Real Memory Layer — Anders Swanson, Oracle
- Designing CLIs for Agents, Not Humans — Pedro Lopez, Airbyte
- Building on the Codex Harness — Dominik Kundel, OpenAI
- Stop Prompting — Greg Pstrucha, Sentry
- Brains vs Hands: How to Run AI Agents Safely in Production — Viren Baraiya