TLDR: This post shows how PyRIT’s executor enables practical multi-agent LLM red teaming by letting one model actively attack another. Using a local Ollama setup, I focus on the Multi-Turn strategy . where attacker, target, and scorer models are orchestrated in a loop, with scoring and stopping logic handled entirely by the executor. The result is a reproducible, extensible way to automate jailbreak discovery. ...
TLDR: Introduces Microsoft Agent Framework (Python/.NET) and focuses on workflow orchestration. Covers agent orchestration without workflows, the motivation for explicit workflows, building blocks of workflows (executors, edges), integration with Dev UI, and how MCP tools can be part of workflows. Demonstrates workflow design (conditional routing, state handling) and contrasts with manual orchestration ...
Focuses on security in MCP. Observes MCP’s expanded attack surface and need for robust auth. Describes updated MCP spec with OAuth 2.1 flows, OAuth resource server classification for MCP, dynamic client registration, PKCE enforcement, resource indicators, and JSON-RPC batching. Shows a proof-of-concept integration of MCP with Entra ID for token verification in FastMCP ...