[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CW39FYBWGC5FXTKF1VXDAQ":3,"skill-same:01M7CW39FYBWGC5FXTKF1VXDAQ":45},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M7CW39FYBWGC5FXTKF1VXDAQ","process-discovery",null,"Map how a business process actually works before deciding what to build. Use when the request is a business outcome rather than a defined feature, when the words used do not match a process you can name, or when asked to discover, map, model, or redesign how work happens. Produces an as-is map, the gaps, and a to-be proposal.","ryasrk\u002Fdsh-expert-agents","ryasrk","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F115061102?v=4","01M7CW079PK1VJJZPP5KXWFRXN","Ryas Rafi Karim","01M7CW07AKNEJV8913PZB163JK",0,"MIT","https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents\u002Fblob\u002Fmain\u002Fassets\u002Fskills\u002Fprocess-discovery\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: process-discovery\ndescription: Map how a business process actually works before deciding what to build. Use when the request is a business outcome rather than a defined feature, when the words used do not match a process you can name, or when asked to discover, map, model, or redesign how work happens. Produces an as-is map, the gaps, and a to-be proposal.\n---\n\n# Process Discovery\n\n## Overview\n\nSoftware is a means to a business end. When a request arrives as an outcome\n(\"reduce order errors\", \"we need an approval flow\", \"customers keep complaining\nabout X\"), there is a process behind it that nobody has written down, and the\ndetails only exist in the heads of the people doing the work.\n\n**Core principle: map what happens before you decide what to build.** A system\nbuilt against the request as stated encodes the requester's mental model, which\nis almost never the process. The most expensive defects are not bugs; they are\nfeatures that faithfully implement a process that was never how the work got\ndone.\n\nThis skill produces three artifacts and stops. It does not choose technology,\ndatabases, or frameworks.\n\n## The four phases\n\nRun them in order. Do not skip to phase 3 because you think you already know.\n\n### Phase 1: Frame the outcome\n\nAnswer, in business language:\n\n- **What outcome does the business want, and how is it measured today?** A\n  metric, not a feeling. \"Time to approve an invoice: 4 days\" is frameable;\n  \"approvals are slow\" is not.\n- **Who carries the pain?** The person who feels the problem is often not the\n  person who requested the change.\n- **What is the boundary?** Which process, which teams, which systems. An\n  unbounded discovery never ends.\n- **What triggered this now?** A regulation, a growth threshold, a lost customer.\n  The trigger tells you which constraint is real.\n\nIf the outcome cannot be stated as something measurable, stop and ask. A\ndiscovery run against an unmeasurable goal produces a model nobody can evaluate.\n\n### Phase 2: Map the as-is process\n\nThis is the phase that pays for the whole exercise. Map what happens, not what\nthe process document says or what the requester believes.\n\nFor each step, capture:\n\n| Field | What to record |\n| --- | --- |\n| Actor | The role that performs it, named as the organisation names it |\n| Trigger | What causes this step to start |\n| Input | What it consumes, and where that came from |\n| Action | What is actually done, including the manual parts |\n| Output | What it produces, and who receives it |\n| Time | How long it takes, and whether that is work time or waiting time |\n| Failure | What happens when it goes wrong, and who notices |\n\nThen walk the map and ask, at every single step: **\"and then what happens?\"**\nFollow every answer. The step after the one the requester described is where the\nreal process usually diverges from the described one.\n\n**Ask about the workarounds specifically.** The spreadsheet beside the system,\nthe side channel, the person who re-checks every entry. A workaround is not a\nuser failing to use the system; it is the system failing to match the process.\nEvery workaround is a requirement nobody has written down yet.\n\n**Ask \"can there be more than one of these at once?\"** of every piece of state.\nA process where one order is open is a different design from one where five are,\nand the difference is invisible until someone tries to represent it.\n\n### Phase 3: Find the gaps\n\nWalk the as-is map and classify each finding:\n\n- **Redundancy:** the same data entered twice, the same check performed by two\n  roles. Evidence: two steps with the same output.\n- **Handoff:** work passes between actors or systems. Each handoff is a queue and\n  a place information is lost. Count them; the count is a defect signal.\n- **Bottleneck:** the step where work waits. Compare work time to waiting time.\n  A step that takes 10 minutes of work but 3 days of waiting is the bottleneck,\n  and it is a queueing problem, not a speed problem.\n- **Workaround:** anything outside the system of record. Each one is a hidden\n  requirement.\n- **Missing feedback:** a step whose failure nobody detects until much later.\n- **Control gap:** a decision with no record of who made it or why, which is what\n  an audit will ask for.\n\n### Phase 4: Propose the to-be\n\nFor each gap, propose a change, and state the **measurable** change it produces.\n\"Automate the approval step, reducing the wait from 3 days to 4 hours\" is a\nproposal. \"Improve the approval flow\" is not.\n\nPrefer eliminating a step to automating it. An automated step that should not\nexist is worse than a manual one, because it now runs faster and produces more.\n\nThen state plainly:\n\n- **What stays manual, and why.** Some steps involve judgment, physical presence,\n  or accountability that a system cannot hold. Naming them prevents the proposal\n  from reading as \"automate everything\".\n- **What must be true before this works.** A dependency on a process change, a\n  data cleanup, or a decision someone has not made.\n- **What this does not fix.** The scope boundary, restated against the findings.\n\n## Output\n\nWrite the artifacts to `docs\u002Fprocess\u002F\u003Cprocess-name>\u002F`:\n\n- `as-is.md`: the phase 2 map, one section per step, with the table above.\n- `gaps.md`: the phase 3 findings, each classified and evidenced.\n- `to-be.md`: the phase 4 proposal, each change with its metric and its cost.\n\nKeep the language the business uses, not what you would name things in code. If\nthe organisation says \"job card\", write \"job card\", even if you would call it an\norder. Renaming things is a phase 6 problem for whoever builds it, and doing it\nhere loses the people who can confirm the map.\n\n## Validation\n\nBefore handing off, check:\n\n- Every step has an actor, a trigger, and an output. A step missing any of the\n  three is a step you have not finished mapping.\n- Every gap finding cites the step or steps that evidence it.\n- Every to-be proposal names a metric it moves.\n- The manual steps that remain are named explicitly.\n- The map uses the business's own vocabulary throughout.\n\n## Handoff\n\nThis skill ends with a mapped process, not a system. It hands off to:\n\n- **domain modeling** when the process names concepts the codebase must share.\n- **planning** once the to-be is agreed, to slice it into executable work.\n- **design** when the to-be implies a boundary or data-shape decision.\n\nDo not begin implementation from this skill's output directly. A to-be proposal\nthat has not been through planning has an unfixed scope and a metric nobody has\nagreed to.\n\n## Anti-patterns\n\n- **Interviewing only the requester.** They know the process they wish existed.\n- **Mapping the documented process.** The document describes the intended\n  process; the workaround describes the actual one.\n- **Jumping to a solution in phase 2.** Once you are proposing a system, you stop\n  asking \"and then what happens?\", and that question is where the value is.\n- **Treating the bottleneck as a speed problem.** Work waiting in a queue is not\n  fixed by making the work faster.\n- **A to-be with no metric.** It cannot be evaluated, so it cannot be rejected,\n  so it will be built.\n","# Process Discovery\n\n## Overview\n\nSoftware is a means to a business end. When a request arrives as an outcome\n(\"reduce order errors\", \"we need an approval flow\", \"customers keep complaining\nabout X\"), there is a process behind it that nobody has written down, and the\ndetails only exist in the heads of the people doing the work.\n\n**Core principle: map what happens before you decide what to build.** A system\nbuilt against the request as stated encodes the requester's mental model, which\nis almost never the process. The most expensive defects are not bugs; they are\nfeatures that faithfully implement a process that was never how the work got\ndone.\n\nThis skill produces three artifacts and stops. It does not choose technology,\ndatabases, or frameworks.\n\n## The four phases\n\nRun them in order. Do not skip to phase 3 because you think you already know.\n\n### Phase 1: Frame the outcome\n\nAnswer, in business language:\n\n- **What outcome does the business want, and how is it measured today?** A\n  metric, not a feeling. \"Time to approve an invoice: 4 days\" is frameable;\n  \"approvals are slow\" is not.\n- **Who carries the pain?** The person who feels the problem is often not the\n  person who requested the change.\n- **What is the boundary?** Which process, which teams, which systems. An\n  unbounded discovery never ends.\n- **What triggered this now?** A regulation, a growth threshold, a lost customer.\n  The trigger tells you which constraint is real.\n\nIf the outcome cannot be stated as something measurable, stop and ask. A\ndiscovery run against an unmeasurable goal produces a model nobody can evaluate.\n\n### Phase 2: Map the as-is process\n\nThis is the phase that pays for the whole exercise. Map what happens, not what\nthe process document says or what the requester believes.\n\nFor each step, capture:\n\n| Field | What to record |\n| --- | --- |\n| Actor | The role that performs it, named as the organisation names it |\n| Trigger | What causes this step to start |\n| Input | What it consumes, and where that came from |\n| Action | What is actually done, including the manual parts |\n| Output | What it produces, and who receives it |\n| Time | How long it takes, and whether that is work time or waiting time |\n| Failure | What happens when it goes wrong, and who notices |\n\nThen walk the map and ask, at every single step: **\"and then what happens?\"**\nFollow every answer. The step after the one the requester described is where the\nreal process usually diverges from the described one.\n\n**Ask about the workarounds specifically.** The spreadsheet beside the system,\nthe side channel, the person who re-checks every entry. A workaround is not a\nuser failing to use the system; it is the system failing to match the process.\nEvery workaround is a requirement nobody has written down yet.\n\n**Ask \"can there be more than one of these at once?\"** of every piece of state.\nA process where one order is open is a different design from one where five are,\nand the difference is invisible until someone tries to represent it.\n\n### Phase 3: Find the gaps\n\nWalk the as-is map and classify each finding:\n\n- **Redundancy:** the same data entered twice, the same check performed by two\n  roles. Evidence: two steps with the same output.\n- **Handoff:** work passes between actors or systems. Each handoff is a queue and\n  a place information is lost. Count them; the count is a defect signal.\n- **Bottleneck:** the step where work waits. Compare work time to waiting time.\n  A step that takes 10 minutes of work but 3 days of waiting is the bottleneck,\n  and it is a queueing problem, not a speed problem.\n- **Workaround:** anything outside the system of record. Each one is a hidden\n  requirement.\n- **Missing feedback:** a step whose failure nobody detects until much later.\n- **Control gap:** a decision with no record of who made it or why, which is what\n  an audit will ask for.\n\n### Phase 4: Propose the to-be\n\nFor each gap, propose a change, and state the **measurable** change it produces.\n\"Automate the approval step, reducing the wait from 3 days to 4 hours\" is a\nproposal. \"Improve the approval flow\" is not.\n\nPrefer eliminating a step to automating it. An automated step that should not\nexist is worse than a manual one, because it now runs faster and produces more.\n\nThen state plainly:\n\n- **What stays manual, and why.** Some steps involve judgment, physical presence,\n  or accountability that a system cannot hold. Naming them prevents the proposal\n  from reading as \"automate everything\".\n- **What must be true before this works.** A dependency on a process change, a\n  data cleanup, or a decision someone has not made.\n- **What this does not fix.** The scope boundary, restated against the findings.\n\n## Output\n\nWrite the artifacts to `docs\u002Fprocess\u002F\u003Cprocess-name>\u002F`:\n\n- `as-is.md`: the phase 2 map, one section per step, with the table above.\n- `gaps.md`: the phase 3 findings, each classified and evidenced.\n- `to-be.md`: the phase 4 proposal, each change with its metric and its cost.\n\nKeep the language the business uses, not what you would name things in code. If\nthe organisation says \"job card\", write \"job card\", even if you would call it an\norder. Renaming things is a phase 6 problem for whoever builds it, and doing it\nhere loses the people who can confirm the map.\n\n## Validation\n\nBefore handing off, check:\n\n- Every step has an actor, a trigger, and an output. A step missing any of the\n  three is a step you have not finished mapping.\n- Every gap finding cites the step or steps that evidence it.\n- Every to-be proposal names a metric it moves.\n- The manual steps that remain are named explicitly.\n- The map uses the business's own vocabulary throughout.\n\n## Handoff\n\nThis skill ends with a mapped process, not a system. It hands off to:\n\n- **domain modeling** when the process names concepts the codebase must share.\n- **planning** once the to-be is agreed, to slice it into executable work.\n- **design** when the to-be implies a boundary or data-shape decision.\n\nDo not begin implementation from this skill's output directly. A to-be proposal\nthat has not been through planning has an unfixed scope and a metric nobody has\nagreed to.\n\n## Anti-patterns\n\n- **Interviewing only the requester.** They know the process they wish existed.\n- **Mapping the documented process.** The document describes the intended\n  process; the workaround describes the actual one.\n- **Jumping to a solution in phase 2.** Once you are proposing a system, you stop\n  asking \"and then what happens?\", and that question is where the value is.\n- **Treating the bottleneck as a speed problem.** Work waiting in a queue is not\n  fixed by making the work faster.\n- **A to-be with no metric.** It cannot be evaluated, so it cannot be rejected,\n  so it will be built.\n",{"name":5,"description":7},"main","b608340412f02546f13b1b29976daf33bec72bef","assets\u002Fskills\u002Fprocess-discovery","https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents","https:\u002F\u002Fgithub.com\u002Fryasrk","Passionate about Machine Learning, Deep Learning, and LLMs. Skilled in Computer Vision, NLP, and AI model development.","2026-09-29 12:47:36",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add ryasrk\u002Fdsh-expert-agents --skill process-discovery",{"tool":37,"command":38},"bunx","bunx skills add ryasrk\u002Fdsh-expert-agents --skill process-discovery",{"tool":40,"command":41},"pnpm","pnpm dlx skills add ryasrk\u002Fdsh-expert-agents --skill process-discovery","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents\u002Fblob\u002Fmain\u002Fassets\u002Fskills\u002Fprocess-discovery\u002FSKILL.md",true,[],[46,53,60,67,74,81],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M7CW5A9FJK1C6BQKKMYHKWBD","writing-shape","Writing, exploit: shape raw material into an article, paragraph by paragraph.","https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents\u002Fblob\u002Fmain\u002Fassets\u002Fskills\u002Fwriting-shape\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M7CW58T8BYGMCTKQW60DFM7K","writing-fragments","Writing, explore: mine raw fragments, no structure yet.","https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents\u002Fblob\u002Fmain\u002Fassets\u002Fskills\u002Fwriting-fragments\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M7CW56PXKJ31Y8AGJ31W1SHT","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents\u002Fblob\u002Fmain\u002Fassets\u002Fskills\u002Fwriting-for-agents\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M7CW54P3DQJKVA9GBQBMXE00","writing-beats","Writing, exploit; assemble raw material into a journey of beats, grounding each term before a beat leans on it.","https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents\u002Fblob\u002Fmain\u002Fassets\u002Fskills\u002Fwriting-beats\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7CW50RQRS31BHTKJ7JZM21A","wizard","Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.","https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents\u002Fblob\u002Fmain\u002Fassets\u002Fskills\u002Fwizard\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M7CW4Y3E85C06AZD2VANK40Q","wayfinder","Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.","https:\u002F\u002Fgithub.com\u002Fryasrk\u002Fdsh-expert-agents\u002Fblob\u002Fmain\u002Fassets\u002Fskills\u002Fwayfinder\u002FSKILL.md",[],[]]