RE: LeoThread 2025-12-31 23-56

A new programmable layer of abstraction must be mastered alongside the usual ones — agents, subagents, prompts, contexts, memory, modes, permissions, tools, plugins, skills, hooks, MCP, LSP, slash commands, workflows, IDE integrations — plus a coherent mental model of the strengths and failure modes of fundamentally stochastic, fallible, opaque, and evolving entities now intertwined with traditional engineering