Fable 5.1 让 Claude Tag 更强大Fable 5.1 Makes Claude Tag Even More Useful
Claude Code 工程师 Boris Cherny 展示 Fable 5.1 如何增强 Claude Tag：它能从 metrics spreadsheet 和跨 Slack 的其他数据即时生成 leadership deck，还能发现与数字不符的 vendor report 并提前标记。Claude Tag 已在 Slack 的 Team 与 Enterprise 计划中可用。
Boris Cherny of Claude Code showed Fable 5.1 making Claude Tag more powerful: it builds a last-minute leadership deck from a metrics spreadsheet and data across Slack, spots a vendor report that disagrees with the numbers, and flags it. Claude Tag is available in Slack on Team and Enterprise plans.
Claude Code engineer Boris Cherny demonstrated how Fable 5.1 makes Claude Tag even more useful. It can build a last-minute leadership deck from a metrics spreadsheet and other data across Slack, spot a vendor report that disagrees with the numbers, and flag it before moving on. Claude Tag is available in Slack on Team and Enterprise plans.
查看原文 →
Compound Writing 插件：让写作像工程一样可复利Compound Writing Plugin: Making Writing Compound Like Engineering
The Takeaway: The real power of AI writing is not one-shot generation but compounding every piece of feedback so the system permanently gets better at your process, voice, and standards.
Katie Parrott, staff writer at Every, has turned two years of hard-won practice into an open plugin called Compound Writing. Inspired by Kieran Klaassen’s Compound Engineering, it treats writing the way good engineering treats code: brainstorm → outline → draft → substantive edit → line edit → final polish, with every correction fed back into the system so the same mistake never needs to be explained twice.
She started as an AI skeptic who used ChatGPT as a $20 career coach after a layoff. That experiment taught her the core lesson she still lives by: “Clarity doesn’t arrive gift-wrapped from a digital assistant… It’s something I had to dig out for myself, question by question and prompt by prompt.” AI became both a productivity multiplier and a supportive technology that reduced friction around anxiety, email, and executive function.
Today her career coach lives inside Codex as a full project with dossiers, OKRs, validation folders, and a Kanban board she never touches. For writing she brings only the fresh, last-mile ingredients—new data, personal insight, current research—while the plugin holds the recipe of structure, voice, and review lenses (Vonnegut story rules, Hitchcock suspense, Sorkin dialogue, Sedaris humor). The result is that writing feels exploratory again instead of a slog.
Her broader thesis is that education and access will matter more than ever. Living in Columbus, Ohio rather than a coastal hub, she sees the risk that AI’s compounding returns concentrate among those who already have time, money, and networks. The compelling future is the one where more people, in more places, get to compound too.
The Takeaway: The real power of AI writing is not one-shot generation but compounding every piece of feedback so the system permanently gets better at your process, voice, and standards.
Katie Parrott, staff writer at Every, has turned two years of hard-won practice into an open plugin called Compound Writing. Inspired by Kieran Klaassen’s Compound Engineering, it treats writing the way good engineering treats code: brainstorm → outline → draft → substantive edit → line edit → final polish, with every correction fed back into the system so the same mistake never needs to be explained twice.
She started as an AI skeptic who used ChatGPT as a $20 career coach after a layoff. That experiment taught her the core lesson she still lives by: “Clarity doesn’t arrive gift-wrapped from a digital assistant… It’s something I had to dig out for myself, question by question and prompt by prompt.” AI became both a productivity multiplier and a supportive technology that reduced friction around anxiety, email, and executive function.
Today her career coach lives inside Codex as a full project with dossiers, OKRs, validation folders, and a Kanban board she never touches. For writing she brings only the fresh, last-mile ingredients—new data, personal insight, current research—while the plugin holds the recipe of structure, voice, and review lenses (Vonnegut story rules, Hitchcock suspense, Sorkin dialogue, Sedaris humor). The result is that writing feels exploratory again instead of a slog.
Her broader thesis is that education and access will matter more than ever. Living in Columbus, Ohio rather than a coastal hub, she sees the risk that AI’s compounding returns concentrate among those who already have time, money, and networks. The compelling future is the one where more people, in more places, get to compound too.
查看原文 →