The last rung of doing everything by hand: spend less, and keep what you've built from quietly getting worse. **Cost and rate-limit engineering.** Why the `/cost` command is meaningful only on a pay-as-you-go API key and misleading on a subscription (use `/status` and `/usage` instead), and the gotcha where a stray `ANTHROPIC_API_KEY` bills you through the API while your Max plan sits unused. The two stacked limit windows (the five-hour rolling window and the seven-day weekly caps, including the separate cap on the top model), drawn from [Anthropic's usage and limits docs](https://support.claude.com/en/articles/14552983-models-usage-and-limits-in-claude-code) and the [Pro/Max plan guide](https://support.claude.com/en/articles/11145838-use-claude-code-with-your-pro-or-max-plan). Current [per-token pricing](https://platform.claude.com/docs/en/about-claude/pricing) and the clean five-times pattern (output is 5x input; each model tier is ~5x cheaper than the one above), prompt caching at a 90% read discount and why a stable CLAUDE.md keeps the cache hot, and the batch path at 50% off. Reading your real usage with [ccusage](https://github.com/ryoppippi/ccusage) and exporting [OpenTelemetry metrics](https://code.claude.com/docs/en/monitoring-usage) to a dashboard. The levers that move the bill most: `/model`, `/compact` and `/clear`, subagents that return summaries, `--max-turns`, and the thinking-budget setting (thinking tokens bill as output). More in [Manage costs effectively](https://code.claude.com/docs/en/costs). **Evaluating your own prompts, skills, and agents.** Why your setup drifts (model updates, CLAUDE.md edits, accumulating instructions) and how regressions stay silent. Building a tiny eval suite with [headless print mode](https://code.claude.com/docs/en/headless): a fixtures folder, a pinned model, and code-based checks (does it compile, do tests pass, does it contain the required clause) before reaching for an LLM-as-judge rubric. [promptfoo](https://github.com/promptfoo/promptfoo) for assertions and judging, Anthropic's [evals guidance](https://github.com/anthropics/claude-cookbooks) that code-based grading wins when feasible, and four ways evals lie to you: tiny overfit sets, judging style over correctness, eval cost, and non-determinism. News up top: Opus 4.8 as the new Claude Code default with extra-high effort and Dynamic Workflows ([docs](https://code.claude.com/docs/en/workflows)), and today's 2.1.160 write-guard prompts ([changelog](https://code.claude.com/docs/en/changelog)). Earlier episodes referenced: permissions and plan mode, custom slash commands, skills, subagents, MCP servers, and context windows.