Tag: agents

All the articles with the tag "agents".

• 95% Agreement, Zero Failures Caught: Testing Your LLM Judge

  Published:Sep 17, 2026 |  at 06:00 AM

  An executable confusion-matrix example shows why agreement can hide a useless judge. Evaluate failure detection and actual task outcomes separately.

• Your MCP Server Is an Authorization Boundary

  Published:Sep 17, 2026 |  at 06:00 AM

  A tool schema is not permission. Separate token audience, client consent, session routing and object ownership in remote MCP services.

• Flue Under the Hood: Why This Agent Harness Holds

  Published:May 16, 2026 |  at 10:00 AM

  A source-level tour of Flue through domain-driven design and twelve-factor architecture: the language, boundaries, and runtime constructs behind the TypeScript agent harness framework.