S01E02 · Theo / Core

Steering Models Through Context

Lesson

Stub

Planning stub — the outline is here; this section is not yet written for learners.

We already know that the quality of an LLM’s output depends heavily on the context available to it: what’s already in its working memory and what it can access easily.

Outline

  • Settings: profiles, projects, agents, permissions, history, connectors
  • Context: sessions, threads, metadata, environment, tools, skills, modes, compression
  • Processing: text, threads, context, multimodality, attachments, real-time interactions
  • Skills: structure, selection, scope, graphs, scripts, dynamic resources, generation
  • Tools: discovery, schemas, results, hints, recovery, code mode, Bash
  • Subagents: mechanics, settings, state, handoffs, shared context, workers

Outcome

Lesson

Takeaways