[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EMBEVEJ8V73RBQVTFF4738":3,"skill-same:01M7EMBEVEJ8V73RBQVTFF4738":47},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":6,"updatedAt":32,"install":33,"resources":46},"01M7EMBEVEJ8V73RBQVTFF4738","automation-builder",null,"Build workflow automations — triggers, actions, conditional logic, error handling, and maintainable automation design.","TanNhatCMS\u002Fjarvis","TanNhatCMS","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113796420?v=4","01M753MV17KG5CD3152G3WQEQY","KietNT","01M7EKEDYVYQ6PWVJJMZ911VMX",0,"MIT","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-automation-builder\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: automation-builder\ndescription: Build workflow automations — triggers, actions, conditional logic, error handling, and maintainable automation design.\ncategory: curviate\n---\n\n## Overview\n\nWorkflow automation connects triggers to actions: when X happens, do Y (with conditions, branching, and error handling). This skill covers designing automations well — whether in no-code builders or code — including trigger selection, action design, conditional logic, testing, monitoring, and the maintenance discipline that keeps automations reliable. Platform-neutral.\n\n\nAutomation builders turn repetitive operational work into reliable systems: lead routing, data syncing, notifications, report generation, and multi-step workflows across SaaS tools.\nThe discipline combines process mapping, tool selection, error handling, and monitoring.\nDone well, automation gives teams leverage — work that scaled linearly with headcount now scales with software.\n## When to use\n\n- Designing workflow automations\n- Choosing triggers and actions\n- Adding conditional logic and branching\n- Debugging failing automations\n- Reducing manual operational work\n- Governing automation sprawl\n\n- Eliminating manual data entry between tools\n- Building lead routing and assignment flows\n- Creating scheduled reports and digest notifications\n- Syncing customer data across platforms\n- Triggering follow-ups from product or CRM events\n## Core concepts\n\n**Trigger design.** Event-based (record created, status changed, webhook received), scheduled (daily digest, weekly cleanup), and manual (button-triggered). Choose the trigger matching the actual need — polling when events exist wastes resources; scheduled batches beat per-event processing for digest-style outputs.\n\n**Action design.** Atomic actions (one thing each — easier to debug and reuse), idempotent where possible (safe to re-run), with clear success\u002Ffailure signals. Chain actions with data passing (outputs → inputs); validate data between steps.\n\n**Conditional logic.** If\u002Fthen branches, filters (only proceed when...), and switch statements for multi-path flows. Keep branches shallow — deeply nested logic becomes unmaintainable. Complex decisions belong in code or decision tables, not 15 nested conditions.\n\n**Error handling.** Per-action: retry transient failures, route permanent failures to alerts, define fallbacks (what happens when the CRM is down?), and set up dead-letter handling. Every automation needs a failure story — \"it just stops\" is not one.\n\n**Testing.** Test with realistic data (not just happy paths), test edge cases (empty results, duplicates, rate limits), dry-run modes before going live, and staged rollouts (one team before all teams). Automations fail at scale in ways tests miss — monitor early.\n\n**Documentation.** Every automation: purpose, trigger, steps, owner, and what to do when it breaks. Undocumented automations become haunted infrastructure nobody dares touch.\n\n\n**Trigger-action logic.** Every automation starts with a trigger (an event: form submitted, deal stage changed, row added) and proceeds through actions (create record, send message, update field).\nGood triggers are specific — \"deal moved to Closed-Won\" beats \"deal updated\" — because vague triggers fire on noise and create side effects.\nMap the trigger's data payload before building: what fields are available? What is missing? Most automation failures trace to assuming data that is not in the payload.\n**Branching and conditions.** Real workflows branch: if deal value > $10k, route to enterprise team; if source = partner, notify partnerships.\nKeep branching shallow (2–3 levels); deep conditional trees become unmaintainable. When logic gets complex, split into multiple focused automations rather than one mega-flow.\n**Error handling.** Every automation needs a failure path: what happens when the API call fails, the field is empty, or the recipient is missing?\nAt minimum: retry transient failures (exponential backoff), alert a human on persistent failure, and log every run with inputs and outcomes.\nSilent failures are the nightmare scenario — the automation stops working and nobody notices for weeks.\n## Practical workflow\n\n1. **Define the job.** What manual work disappears? What's the trigger, what's the outcome, who owns it? If you can't describe it in one sentence, it's not ready to automate.\n2. **Map the flow.** Trigger → steps → branches → error paths. Whiteboard first; identify: data needed at each step, decision points, and failure modes.\n3. **Build simply.** Start with the happy path + basic error handling. Atomic actions, shallow branching, idempotent writes. Resist cleverness.\n4. **Test thoroughly.** Happy path, edge cases (empty, duplicate, malformed), failure injection (what if step 3 fails?), and load (what if 1,000 events fire at once?).\n5. **Deploy carefully.** Dry-run or shadow mode first, then limited rollout, then full. Notify affected humans (\"this is now automated; here's what changed\").\n6. **Monitor and maintain.** Run history review, failure alerts, performance tracking, and quarterly audits (is this automation still needed? still correct?). Assign owners — orphaned automations rot.\n\n**Automation spec template:** name → owner → purpose (one line) → trigger → steps → conditions → error handling → monitoring → review date.\n\n\n**Build process:** 1) document the manual process as-is (steps, actors, time spent), 2) identify the trigger and required data, 3) design the happy path, 4) design failure paths, 5) build in a sandbox, 6) test with real data (10+ cases including edge cases), 7) soft-launch with monitoring, 8) document and hand over.\n**Monitoring dashboard:** runs per day, success rate, average duration, failure alerts, and data-quality flags.\nReview weekly for the first month, monthly after. Automations rot as tools change — monitoring catches the rot.\n**Documentation template:** purpose → trigger → steps → expected outcomes → failure handling → owner → last reviewed date.\nIf the owner leaves and nobody can understand the automation, it is a liability, not an asset.\n## Common pitfalls\n\n- **Automating broken processes.** Speeding up a bad workflow. Fix the process first, then automate.\n- **No error handling.** Happy-path-only automations that silently fail. Every step needs a failure plan.\n- **Over-complexity.** 50-step mega-flows. Break into smaller, composable automations.\n- **No ownership.** Automations nobody maintains. Every automation has a named owner and review date.\n- **Trigger spam.** Per-event processing for digest needs. Match trigger type to the actual requirement.\n- **Untested edge cases.** Works for 10 records, breaks at 10,000. Test scale and weirdness.\n- **Shadow IT sprawl.** Dozens of personal automations with no visibility. Central inventory and governance for shared systems.\n- **Automating broken processes.** Speeding up a bad workflow just produces bad outcomes faster. Fix the process first, then automate.\n- **No ownership.** Automations with no named owner break silently. Every automation has an owner and a review date.\n- **Over-automation.** Automating judgment calls that need humans (complex negotiations, sensitive communications). Automate the routine; keep humans on the exceptions.\n- **Circular triggers.** Automation A updates a record, triggering automation B, which triggers A again. Map trigger chains before building; add guards against loops.\n","## Overview\n\nWorkflow automation connects triggers to actions: when X happens, do Y (with conditions, branching, and error handling). This skill covers designing automations well — whether in no-code builders or code — including trigger selection, action design, conditional logic, testing, monitoring, and the maintenance discipline that keeps automations reliable. Platform-neutral.\n\n\nAutomation builders turn repetitive operational work into reliable systems: lead routing, data syncing, notifications, report generation, and multi-step workflows across SaaS tools.\nThe discipline combines process mapping, tool selection, error handling, and monitoring.\nDone well, automation gives teams leverage — work that scaled linearly with headcount now scales with software.\n## When to use\n\n- Designing workflow automations\n- Choosing triggers and actions\n- Adding conditional logic and branching\n- Debugging failing automations\n- Reducing manual operational work\n- Governing automation sprawl\n\n- Eliminating manual data entry between tools\n- Building lead routing and assignment flows\n- Creating scheduled reports and digest notifications\n- Syncing customer data across platforms\n- Triggering follow-ups from product or CRM events\n## Core concepts\n\n**Trigger design.** Event-based (record created, status changed, webhook received), scheduled (daily digest, weekly cleanup), and manual (button-triggered). Choose the trigger matching the actual need — polling when events exist wastes resources; scheduled batches beat per-event processing for digest-style outputs.\n\n**Action design.** Atomic actions (one thing each — easier to debug and reuse), idempotent where possible (safe to re-run), with clear success\u002Ffailure signals. Chain actions with data passing (outputs → inputs); validate data between steps.\n\n**Conditional logic.** If\u002Fthen branches, filters (only proceed when...), and switch statements for multi-path flows. Keep branches shallow — deeply nested logic becomes unmaintainable. Complex decisions belong in code or decision tables, not 15 nested conditions.\n\n**Error handling.** Per-action: retry transient failures, route permanent failures to alerts, define fallbacks (what happens when the CRM is down?), and set up dead-letter handling. Every automation needs a failure story — \"it just stops\" is not one.\n\n**Testing.** Test with realistic data (not just happy paths), test edge cases (empty results, duplicates, rate limits), dry-run modes before going live, and staged rollouts (one team before all teams). Automations fail at scale in ways tests miss — monitor early.\n\n**Documentation.** Every automation: purpose, trigger, steps, owner, and what to do when it breaks. Undocumented automations become haunted infrastructure nobody dares touch.\n\n\n**Trigger-action logic.** Every automation starts with a trigger (an event: form submitted, deal stage changed, row added) and proceeds through actions (create record, send message, update field).\nGood triggers are specific — \"deal moved to Closed-Won\" beats \"deal updated\" — because vague triggers fire on noise and create side effects.\nMap the trigger's data payload before building: what fields are available? What is missing? Most automation failures trace to assuming data that is not in the payload.\n**Branching and conditions.** Real workflows branch: if deal value > $10k, route to enterprise team; if source = partner, notify partnerships.\nKeep branching shallow (2–3 levels); deep conditional trees become unmaintainable. When logic gets complex, split into multiple focused automations rather than one mega-flow.\n**Error handling.** Every automation needs a failure path: what happens when the API call fails, the field is empty, or the recipient is missing?\nAt minimum: retry transient failures (exponential backoff), alert a human on persistent failure, and log every run with inputs and outcomes.\nSilent failures are the nightmare scenario — the automation stops working and nobody notices for weeks.\n## Practical workflow\n\n1. **Define the job.** What manual work disappears? What's the trigger, what's the outcome, who owns it? If you can't describe it in one sentence, it's not ready to automate.\n2. **Map the flow.** Trigger → steps → branches → error paths. Whiteboard first; identify: data needed at each step, decision points, and failure modes.\n3. **Build simply.** Start with the happy path + basic error handling. Atomic actions, shallow branching, idempotent writes. Resist cleverness.\n4. **Test thoroughly.** Happy path, edge cases (empty, duplicate, malformed), failure injection (what if step 3 fails?), and load (what if 1,000 events fire at once?).\n5. **Deploy carefully.** Dry-run or shadow mode first, then limited rollout, then full. Notify affected humans (\"this is now automated; here's what changed\").\n6. **Monitor and maintain.** Run history review, failure alerts, performance tracking, and quarterly audits (is this automation still needed? still correct?). Assign owners — orphaned automations rot.\n\n**Automation spec template:** name → owner → purpose (one line) → trigger → steps → conditions → error handling → monitoring → review date.\n\n\n**Build process:** 1) document the manual process as-is (steps, actors, time spent), 2) identify the trigger and required data, 3) design the happy path, 4) design failure paths, 5) build in a sandbox, 6) test with real data (10+ cases including edge cases), 7) soft-launch with monitoring, 8) document and hand over.\n**Monitoring dashboard:** runs per day, success rate, average duration, failure alerts, and data-quality flags.\nReview weekly for the first month, monthly after. Automations rot as tools change — monitoring catches the rot.\n**Documentation template:** purpose → trigger → steps → expected outcomes → failure handling → owner → last reviewed date.\nIf the owner leaves and nobody can understand the automation, it is a liability, not an asset.\n## Common pitfalls\n\n- **Automating broken processes.** Speeding up a bad workflow. Fix the process first, then automate.\n- **No error handling.** Happy-path-only automations that silently fail. Every step needs a failure plan.\n- **Over-complexity.** 50-step mega-flows. Break into smaller, composable automations.\n- **No ownership.** Automations nobody maintains. Every automation has a named owner and review date.\n- **Trigger spam.** Per-event processing for digest needs. Match trigger type to the actual requirement.\n- **Untested edge cases.** Works for 10 records, breaks at 10,000. Test scale and weirdness.\n- **Shadow IT sprawl.** Dozens of personal automations with no visibility. Central inventory and governance for shared systems.\n- **Automating broken processes.** Speeding up a bad workflow just produces bad outcomes faster. Fix the process first, then automate.\n- **No ownership.** Automations with no named owner break silently. Every automation has an owner and a review date.\n- **Over-automation.** Automating judgment calls that need humans (complex negotiations, sensitive communications). Automate the routine; keep humans on the exceptions.\n- **Circular triggers.** Automation A updates a record, triggering automation B, which triggers A again. Map trigger chains before building; add guards against loops.\n",{"name":5,"category":24,"description":7},"curviate","main","af388a77c2fe5e0b4fd879cca4300a0ffa9a6579","skills\u002Fams-automation-builder",1,"https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis","https:\u002F\u002Fgithub.com\u002FTanNhatCMS","https:\u002F\u002Fmrkiet.name.vn","2026-11-07 14:23:08",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add TanNhatCMS\u002Fjarvis --skill automation-builder",{"tool":39,"command":40},"bunx","bunx skills add TanNhatCMS\u002Fjarvis --skill automation-builder",{"tool":42,"command":43},"pnpm","pnpm dlx skills add TanNhatCMS\u002Fjarvis --skill automation-builder","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-automation-builder\u002FSKILL.md",true,[],[48,55,63,70,77,84],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7EWYJR0J8DARY7E59Z6Y9MV","web-scrape","Turn web pages into LLM-ready markdown via the Firecrawl API (renders JS, anti-bot handling). Use when browser_open\u002Ffetch fails on JS-heavy pages, or when you need clean markdown\u002Fstructured extraction from a site. Requires FIRECRAWL_API_KEY.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fweb-scrape\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":59,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7EWYBN5P6F2AJG3FR24T9BP","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\u002FCSS, 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.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7EWY8FQFNQ2GB7NZZWGSSDX","test-driven-development","Use when implementing any feature or bugfix, before writing implementation code","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-tdd\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7EWY5SHBFRH1HDHVWM9K7TV","systematic-debugging","Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-debugging\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7EWY3NM0Z0JAKS7HDZ24K66","requesting-code-review","Use when completing tasks, implementing major features, or before merging to verify work meets requirements","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-code-review\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7EWY0WRMWX4M9ZFY0WKC593","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).","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-brainstorming\u002FSKILL.md",[],[]]