• MCP — The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.

If you branch on MCP tool failures, isError plus typed fields will not tell you whether to retry, reauthenticate, fix an argument, or switch tools. Recovery logic ends up interpreting prose, which is the fragile path.

• articleCallability Is Not Operability: Controlled Interface Interventions for LLM AgentsZihao Wang
• articletrajectory-judge: What Outcome-Only LLM Judges Miss on Agent TrajectoriesHadi Mohammadi
• articleWhen May an Agent Stop? Evidence-Carrying Termination for Tool-Using LLMsJason Liu

Checking sign-in…

Loading comments…