# Productivity Playbook — Template & Tone Guide The exact section structure every `-Productivity-Playbook.md` must follow. Read this during Step 5 of the parent SKILL.md. --- ## File header (always identical shape) ```markdown # Productivity Playbook > **How people are actually using to run their lives / ship code / operate businesses.** > Last updated: . Research window: → . > Companion files: [other four playbooks] --- ``` --- ## Required sections (in order) Do not renumber these across runs. Month-over-month diffs depend on stable numbering. ### 0. Why `` is different from the other four One-table comparison vs. the sibling tools. Always 2–4 rows max. Example axes: interface, runtime, execution layer, orchestration role, registry scale. ### 1. Foundation — what is `` + install + file anatomy - Origin, maintainer, scale signals (stars, install counts, notable adopters) - Install command(s) — copy-pasteable - File / config anatomy — what lives where on disk ### 2. What's new since `` **This section is the point of monthly re-runs.** Bullet list of: - Product / feature announcements from the vendor - Notable new skills / extensions / plugins - Community workflows that gained traction - Deprecations or breaking changes If empty, say so honestly. Don't pad. ### 3. The flagship workflow / killer pattern The single most-cited or most-copied workflow for this tool. Steinberger's 6,600-commit/month pattern for OpenClaw, Composer 2's parallel agents for Cursor, Codex CLI subagents, Gemini Deep Research Max, Claude's Artifacts + Skills combo, etc. Pick the one that's **actually being copied on X**, not the one the vendor markets hardest. ### 4–11. Persona sections (all eight, in this order) See [`personas.md`](./personas.md). Each persona gets: - 2–5 concrete workflows with links - Real names / handles / case studies when available - A table or bullet list — not a paragraph The eight personas are fixed: 4. PhD / research students 5. Solopreneur / startup founder 6. Marketer / growth / agency 7. Designer (app / web / brand) 8. Video / podcast / creator 9. Software developer 10. Personal productivity / PKM / student 11. Sales / finance / ops / back-office If a persona has nothing meaningful this month, write one honest paragraph and move on. Do not skip the heading — that breaks consistency. ### 12. Unusual / frontier use cases 3–8 genuinely weird examples from X / Reddit / blogs. The purpose is to show the range of the tool, not to pad. Aim for stories that are either: - High-impact (car purchase negotiation, $X reclaimed, millions earned) - Surprising (AI VTuber, agent imageboard, theology bot) - Technically novel (cross-tool orchestration, new primitive) ### 13. Compared to the other four tools in this series One table mapping scenario → best tool. Be honest about where `` loses. ### 14. Four-week getting-started plan Broken into Week 1 / Week 2 / Week 3 / Week 4 with concrete installs and practices. Do not write "explore the docs" — write "install X, Y, Z and build workflow A". ### 15. Safety & security Non-optional for tools with system access (OpenClaw, Cursor with shell, Codex in agent mode, Claude Code, Gemini CLI). - Trust model - Sandboxing / approval options - Audit tooling - Cost guardrails - Specific red-flag skills/behaviors to avoid ### 16. Master link index Grouped: - Official — vendor docs / homepage / GitHub org - Community — awesome lists, marketplaces, directories - Setup guides — by channel / OS / scenario - Best articles / interviews — editorial picks from the research window - Deep how-to content — linked guides Every link must be canonical and verified-live. ### 17. Final take Short. 3–5 sentences. Honest assessment: - Who is this tool actually for? - When would another tool be better? - One-line positioning vs. the other four. Then a pointer to the sibling Skills Catalog: `See the companion file — [-Skills-Catalog.md](./-Skills-Catalog.md) — for the rated, persona-sorted skill install list.` --- ## Tone rules - Third person. Never "you can…" or "we recommend…" — use "people use it to…" / "this is the pattern that…" - Every factual claim cites a link in parentheses or via Markdown link. - Tables > bullets > prose. In that order. - Concrete verbs. Banned: "leverage", "unlock", "empower", "revolutionize", "seamlessly", "cutting-edge". - No emojis in body text. Rating badges (✅🟢🟡🔴★) are fine in catalog files, not here. - Numbers > adjectives: "6,600 commits in January" not "highly productive". - Names > roles: "Peter Steinberger" not "the founder". --- ## Length guide | Section | Target lines | |---|---| | Header + section 0 | 20–40 | | Sections 1–3 | 40–80 each | | Persona sections (4–11) | 30–80 each | | Section 12 (unusual) | 30–60 | | Sections 13–17 | 30–80 each | | **Total body** | **300–700 lines** | Longer than 750 = padding. Shorter than 250 = insufficient research. --- ## Diff-friendly formatting Since this file will be regenerated monthly: - Keep heading text identical run-to-run (`## 5. Solopreneur / startup founder`, not `## 5. Solopreneurs & Founders`). - Date stamps always `YYYY-MM-DD`. - Link lists in alphabetical order within each subsection. - Table rows in stable order (alphabetical by skill name or by descending usefulness rating). This makes `git diff` between monthly versions actually useful for the human reader.