ams-rote
暂无简介
- License 许可
- MIT
- 最近更新
- 2026-11-07
- Stars
- ★ 0
- Forks
- 1
03 SKILL.md 正文
Rote
Overview
Rote work = repetitive, low-judgment tasks done the same way every time: data entry, reformatting, status copying, manual checks.
It's the highest-ROI automation target: frequent, well-defined, and currently eating human hours that could create value.
The rote method: inventory it, measure it, then eliminate (automate, delegate, or delete) — in that order of preference.
When to use
- Days filled with repetitive manual tasks
- Teams doing the same data shuffling weekly
- Finding automation opportunities systematically
- Building a personal or team leverage habit
- Deciding what to automate first
Core concepts
- Rote inventory. For two weeks, log every repetitive task: what, how often, how long, how error-prone. The inventory is the opportunity map.
- Cost math. Frequency x duration x (1 + error cost). A 10-min daily task = 40+ hrs/year. Rote work is expensive in aggregate — make the cost visible.
- Eliminate first. Before automating: can it be deleted? Can the process change so the task vanishes? The best automation is elimination.
- Automate second. If it must exist: script it, template it, or flow it. Match tool to complexity; timebox the build.
- Delegate third. If it needs a human but not you: delegate with a checklist/SOP. Delegation without documentation just moves the pain.
- SOPs for the remainder. Rote work that stays manual gets a checklist/SOP: consistent, delegable, improvable. Undocumented rote is fragile rote.
- Leverage habit. Weekly: 'what did I do repetitively this week?' Monthly automation hour. Leverage compounds like interest.
- Error cost. Rote work isn't just slow — it's error-prone. Factor mistake cost (wrong data sent, missed step) into the automation case.
Practical workflow
- Run the inventory. Two weeks logging repetitive tasks: task, frequency, minutes, errors. No judgment, just data.
- Do the cost math. Annualize each: frequency x duration. Sort by cost. The top 5 are your automation backlog.
- Eliminate. For each top item: can it die? Challenge the process — many rote tasks exist from inertia, not need.
- Automate the survivors. Pick the top 2-3. Timebox builds. Ship the simple version; iterate only if it earns it.
- Document the rest. Remaining manual rote gets SOPs/checklists. 10 minutes documenting saves hours of re-teaching.
- Build the habit. Weekly 15-min review: new rote spotted? Monthly automation hour: kill one item from the backlog.
- Share wins. Show the team what got eliminated and the hours saved. Leverage culture spreads by example.
- Re-inventory quarterly. New rote creeps in constantly. Quarterly inventory keeps the backlog honest and the habit alive.
Common pitfalls
- Automating before eliminating. Building elaborate automation for a task that could simply stop. Delete first, always.
- No inventory. 'Automate things' without knowing what the rote work actually is. The inventory directs effort to real costs.
- Gold-plating. Spending 20 hours automating a 5-min monthly task. Cost math first; timebox the build.
- Automating judgment work. Rote = low judgment. Automating decisions that need human nuance creates confident, fast mistakes.
- Undocumented delegation. Handing rote to someone without SOPs. Delegation without documentation is just relocated suffering.
- One-time automation theater. Automating something done twice a year. The inventory's frequency column prevents this.
- Ignoring error costs. Counting only time saved, not mistakes prevented. Error reduction often exceeds time savings in value.
- Leverage guilt. Feeling bad about 'not doing real work' while automating. Eliminating rote IS real work — it's leverage.
04 同仓库更多 Skills
web-scrape
Turn web pages into LLM-ready markdown via the Firecrawl API (renders JS, anti-bot handling). Use when browser_open/fetch fails on JS-heavy pages, or when you need clean markdown/structured extraction from a site. Requires FIRECRAWL_API_KEY.
design-taste-frontend
tasteskill v2, the anti-slop frontend skill. Art direction for websites, distilled from a study of 247 curated gallery sites (One Page Love, Awwwards Sites of the Day). Deals researched directions (idea, composition, type, palette, photographs, section blueprints) from a library, enforces a strict no-slop house style with a real-browser inspector, and runs a look-and-fix loop on screenshots. Use this whenever you build, design, redesign, restyle, polish or review a landing page, marketing site, portfolio, product page, one-pager, hero or any public-facing web page, in any stack (HTML/CSS, React, Next.js, Vue, Svelte, Astro, Tailwind) — also when the user says a page looks generic, templated, "AI-made", sloppy, or asks to make it look designed, premium, award-level or less like every other site. Not for dashboards or app screens.
test-driven-development
Use when implementing any feature or bugfix, before writing implementation code
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
requesting-code-review
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).