Posts tagged Ai-agents.
Where the first cases come from, which ones earn a place in the suite, why every case that pushes a behavior needs one that bounds it, how to grade without an LLM judge, and what the first honest run surfaced.
Anthropic deleted over 80% of Claude Code system prompt for Claude 5 with no measurable loss on their coding evals. The instructions did not become wrong. The model outgrew needing them. Here is the maintenance pass that finds which of yours have expired, and why you need evals before you delete a single line.
Agent Plugins is an open standard for packaging Agent Skills and MCP servers into one portable folder any client can load. What plugin.json and mcp.json actually define, which clients support it, and whether to package one yet.
Write AI agent evals that catch what you missed: source cases from production traces, spot tests that pass for the wrong reason, and know when to overrule an LLM judge.
What changes for OAuth when you move an MCP server to v2. The MCP TypeScript v2 SDK moves the auth helpers into @modelcontextprotocol/express and surfaces auth info via ctx.http.authInfo — your Keycloak setup stays exactly the same.