Deep_Dives
Short how-to guidelines and long-form deep dives. High information density, system-architecture breakdowns, and hard numbers.
[ENTRIES: 7]Claude Code /goal: keep a session running until a fresh model says it's done
/goal sets a completion condition and Claude keeps taking turns until a separate small fast model (default Haiku) confirms it from the transcript. Requires v2.1.139+, one goal per session, condition up to 4,000 characters.
Claude Code plan mode: explore before you edit, and what it really enforces
Plan mode makes Claude Code read and plan without touching disk until you approve — enable it with Shift+Tab or --permission-mode plan. Under the hood it's prompt reinforcement plus a markdown plan file, not a hard tool lockdown.
How Claude Code's agentic loop fills your context window
Claude Code runs a gather→act→verify loop while startup load, file reads, and tool output fill context before you hit send — compaction, /clear, and subagents are the official levers for what sticks.
Pick the right Claude tier for agent workflows
Opus 4.8 orchestrates, Haiku 4.5 does the parallel grunt work, Sonnet 4.6 is the default in between. Choose by the job and the token bill, not the benchmark.
Claude Sonnet 5 closes the gap to Opus without the Opus bill
Sonnet 5 ships as the default on Free and Pro with intro API pricing at $2/$10 per million tokens — near Opus 4.8 on agentic benchmarks at roughly 40% of Opus input cost when you keep effort at Medium or below.
Context engineering beats a bigger context window
Claude Code runs up to a 1M-token window on flagship models, but a bigger window doesn't fix a cluttered one — every turn replays the whole history and quality drops before the cap.
Why agents ignore your CLAUDE.md
A bloated CLAUDE.md gets filtered out as noise. Keep it to universal instructions and move everything else behind progressive disclosure.