<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>CodeTwig</title><description>Claude Code recipes that are run before they are published.</description><link>https://codetwig.dev/</link><item><title>What a short CLAUDE.md actually changes: 3 runs with, 3 without</title><link>https://codetwig.dev/recipes/claude-md-that-changes-behavior/</link><guid isPermaLink="true">https://codetwig.dev/recipes/claude-md-that-changes-behavior/</guid><description>We gave Claude Code the same task six times. Without a CLAUDE.md it edited a generated file in 2 of 3 runs. With seven lines of rules, 0 of 3, and it was cheaper too.</description><pubDate>Tue, 13 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Unrelated tasks, same answer on /clear</title><link>https://codetwig.dev/recipes/clear-between-unrelated-tasks/</link><guid isPermaLink="true">https://codetwig.dev/recipes/clear-between-unrelated-tasks/</guid><description>Our last test found staying in one session cheaper for related tasks. We repeated it with three unrelated tasks in different parts of a codebase. Staying still won on cost, with one catch.</description><pubDate>Tue, 13 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Is /clear worth it? Not for small, related tasks</title><link>https://codetwig.dev/recipes/keep-context-clean/</link><guid isPermaLink="true">https://codetwig.dev/recipes/keep-context-clean/</guid><description>Three small tasks in the same files, run in one long session or with a fresh session each time. Same results, but clearing cost more, because the long session&apos;s context was cached.</description><pubDate>Tue, 13 Oct 2026 00:00:00 GMT</pubDate></item><item><title>How to set up a Claude Project so it works from the first message</title><link>https://codetwig.dev/recipes/set-up-a-claude-project/</link><guid isPermaLink="true">https://codetwig.dev/recipes/set-up-a-claude-project/</guid><description>What to link, what to write in the instructions, and what to hand over on day one, written by a Claude that works inside one. Each step says what you do and what Claude needs from you.</description><pubDate>Tue, 13 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Hook, skill or CLAUDE.md line? We ran all three (and nothing)</title><link>https://codetwig.dev/recipes/skills-subagents-hooks-commands/</link><guid isPermaLink="true">https://codetwig.dev/recipes/skills-subagents-hooks-commands/</guid><description>We asked Claude Code to follow a house formatting rule four ways, 3 runs each. All four worked. The cheapest was one CLAUDE.md line, and the hook cost more than we expected.</description><pubDate>Tue, 13 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Subagents for codebase searches cost more and saved nothing</title><link>https://codetwig.dev/recipes/subagents-for-big-searches/</link><guid isPermaLink="true">https://codetwig.dev/recipes/subagents-for-big-searches/</guid><description>A three-step refactor in a 24-file API, with and without a CLAUDE.md line telling Claude to delegate searches to a subagent. Both got everything right. Delegating cost more, at least at this size.</description><pubDate>Tue, 13 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Haiku was enough</title><link>https://codetwig.dev/recipes/which-model-for-coding/</link><guid isPermaLink="true">https://codetwig.dev/recipes/which-model-for-coding/</guid><description>We gave all three models a checkout feature with coupons, tax and edge cases. All nine runs passed every hidden test. Haiku averaged under a cent; Opus cost 25 times more.</description><pubDate>Tue, 13 Oct 2026 00:00:00 GMT</pubDate></item></channel></rss>