[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DDBD2G9WVYVFCTYNYFTWAE":3,"skill-same:01M7DDBD2G9WVYVFCTYNYFTWAE":44},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":13,"repositoryHtmlUrl":26,"repositoryHomepageUrl":6,"ownerHtmlUrl":27,"ownerBlogUrl":6,"ownerDescription":28,"updatedAt":29,"install":30,"resources":43},"01M7DDBD2G9WVYVFCTYNYFTWAE","poc",null,"Build, run and evaluate a throwaway Go spike that answers one falsifiable question. Produces a POC Report with verbatim evidence. Keywords: POC, proof of concept, spike, prototype, prove it, experiment, try it out.","girishmotwani\u002Fatom","girishmotwani","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F20937791?v=4","01M7DD9G6RCNXB9G6A33Z0GA1P","01M7DD9G7VT0Z0JHV97PSR0GN9",0,"MIT","https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom\u002Fblob\u002Fmain\u002Ftools\u002Fatom\u002Fassets\u002Fskills\u002Fpoc\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: poc\ndescription: 'Build, run and evaluate a throwaway Go spike that answers one falsifiable question. Produces a POC Report with verbatim evidence. Keywords: POC, proof of concept, spike, prototype, prove it, experiment, try it out.'\n---\n\n# \u002Fatom:poc\n\nThe **PROVE** beat. Builds a disposable Go program that answers exactly one question,\nruns it, and writes up what actually happened.\n\nUsed twice in the pipeline: once for the project, then once per phase that carries\ntechnical risk.\n\n## When to use\n\n- A POC Brief exists and has been approved\n- A phase in the plan has an open technical risk\n- Any moment someone is about to design around an unverified assumption\n\n## Writing artifacts\n\nSend complete POC reports, findings, decisions, and throwaway spike documentation as\nUTF-8 standard input to `atom-write-artifact \u003Crelative-path>`. Treat any refusal or\nnon-zero exit as a blocker; never bypass it with a generic editor or another shell\ncommand.\n\n## Before you start\n\nRead, in this order:\n\n1. [poc-discipline.md](..\u002Fatom\u002Freferences\u002Fpoc-discipline.md) — **the authority for this skill**\n2. [anti-rationalization.md](..\u002Fatom\u002Freferences\u002Fanti-rationalization.md)\n3. [verification.md](..\u002Fatom\u002Freferences\u002Fverification.md)\n4. [treatment-evidence.md](..\u002Fatom\u002Freferences\u002Ftreatment-evidence.md) — **required if the spike compares arms**\n\nYou need an approved brief. If there isn't one, go back to `\u002Fatom:discover`. Building a\nspike without an agreed question produces a program nobody can grade.\n\n---\n\n## Step 1 — Restate the question\n\nCopy the question out of the brief **verbatim** into your working notes.\n\nDo not reword it. Rewording is how a miss becomes a hit: the question drifts toward\nwhatever the spike turned out to be good at, and the report confirms something nobody\nasked.\n\n## Step 2 — Scaffold\n\n```bash\natom poc new {slug}\n```\n\nCreates `poc\u002F{slug}\u002F` with its own `go.mod`, a `main.go` stub, and a `README.md`\ncarrying the question. The spike is a **separate module** — it must not compile into\nthe production build, and it must be free to pull dependencies that production will\nnever take.\n\n## Step 3 — Write the spike\n\nWrite the smallest program that can answer the question.\n\n**This code is disposable, and should look it.** Hardcode values. Skip error handling.\n`panic` on anything unexpected. Put it all in `main.go`. Time spent making a spike\nelegant is time spent on code you are about to delete.\n\nThe one thing a spike must do well is **report**. It should print what it found,\nclearly enough that the output alone settles the question:\n\n```go\n\u002F\u002F The spike's entire job is to answer one question, so it prints the measurement\n\u002F\u002F the question asks about — not a generic \"done\".\nfmt.Printf(\"processed=%d duration=%s p99=%s\\n\", processed, elapsed, p99)\n```\n\nIf you find yourself building abstractions, splitting files, or adding configuration,\nthe question is too broad. Stop, narrow it, and check the narrowing with the developer.\n\n### The scope trap\n\nSpikes grow. You will be tempted to also check the adjacent thing while you're in\nhere. Don't. A spike that answers three questions answers none of them cleanly, and\nits report cannot be graded. Note the adjacent question in `findings.md` as a\ncandidate for a later spike.\n\n## Step 4 — Run it\n\n```bash\natom poc run {slug}\n```\n\nRuns `go vet`, `go build`, then `go run`, and captures **all** output.\n\n> **A POC that has not been executed does not exist.**\n\nReading the spike and concluding it would work is precisely the failure this whole\nframework is built to prevent. If it doesn't run, it has proven nothing.\n\nCapture the real output. You will paste it verbatim — never retyped, never rounded,\nnever reconstructed from memory. If you lost it, run it again.\n\nIf it measures something, run it more than once. A single timing sample is noise.\n\n### Comparison spikes: prove the treatment was applied\n\nSkip this if the spike has one arm. It is mandatory the moment the spike compares\narms and intends to credit the difference to a treatment.\n\nWork through the three checks in\n[treatment-evidence.md](..\u002Fatom\u002Freferences\u002Ftreatment-evidence.md) **before reading any\nresult**:\n\n| # | Check | How |\n|---|---|---|\n| 1 | The treatment reached the model | Render each arm's prompt, compare sizes |\n| 2 | Any required action actually happened | Count entries in the treatment's own usage log |\n| 3 | The treated arm behaved differently at all | Compare a behavioural measure against control |\n\nA treatment carried in the prompt arrives by construction and check 1 settles it. A\ntreatment the agent has to **do something** to use — run a script, query an index,\ncall an API — can fail silently, completing the run with no error while measuring\nnothing. Those treatments must log their own use.\n\n> **The agent's account of what it did is not evidence.** An agent that was blocked\n> may describe the action in the past tense and produce output in the right shape for\n> a tool it never ran.\n\nIf a check fails, the arm has proven nothing. Fix the cause, record the correction as\na deviation, and rerun — do not reinterpret the numbers you already have.\n\n## Step 5 — Reach a verdict\n\nThree outcomes. **All three are successes.**\n\n| Verdict | Meaning | Next |\n|---|---|---|\n| **Confirmed** | The assumption held, and the output shows it | Proceed to `\u002Fatom:spec` |\n| **Refuted** | The assumption was wrong | Back to `\u002Fatom:discover` with what you learned |\n| **Inconclusive** | The spike didn't isolate the question | Sharpen and respike. Do **not** proceed |\n\nGrade against the brief's success criteria, not against how you feel about the code.\n\n**Inconclusive is not a soft confirm.** If the spike crashed, or the numbers came out\nambiguous, or you only got it working by changing what you were testing — the verdict\nis Inconclusive. Writing that up as Confirmed launders a guess into the design as\nevidence, and every decision downstream inherits it.\n\n> Refuting your own plan is the highest-value outcome in this pipeline. You just saved\n> the design. Report it with enthusiasm, not apology.\n\n## Step 6 — Write the POC Report\n\nFill in [poc-report.md](..\u002Fatom\u002Ftemplates\u002Fpoc-report.md) at\n`specs\u002F{project}\u002Fpoc-report.md`, or `specs\u002F{project}\u002Fphase-NN-{slug}\u002Fpoc-report.md`\nfor a phase spike.\n\nTwo sections carry most of the value:\n\n- **Evidence** — verbatim output, in fenced blocks.\n- **Explicitly Not Covered** — the boundary of what this proves. A spike showing a\n  happy path at 1k req\u002Fs is not evidence about error paths at 100k req\u002Fs, and the\n  report must say so, or somebody will later believe it was.\n\nRecord anything that surprised you. Surprises are the highest-value content in the\nreport — they are, by definition, the things the design would otherwise have got wrong.\n\nAppend to `findings.md` **as you go**, not from memory afterwards.\n\n## Step 7 — Gate\n\nPresent per [human-gates.md](..\u002Fatom\u002Freferences\u002Fhuman-gates.md): the verdict, the\nevidence, what it changes, and one specific question.\n\n**End your turn. Wait.**\n\n---\n\n## After approval\n\nLeave the spike in `poc\u002F{slug}\u002F`. It is documentation now — the next person who\ndoubts the decision can re-run it.\n\n**Do not copy spike code into the implementation.** It earned its shortcuts by being\ndisposable. Production code earns its shape from the spec and the tests. Carry the\nknowledge forward, not the source.\n\n---\n\n## Done when\n\n- [ ] Question restated verbatim from the brief\n- [ ] Spike scaffolded as its own module under `poc\u002F{slug}\u002F`\n- [ ] Spike **executed**, with real output captured\n- [ ] For comparison spikes: treatment evidence collected for every arm before results were read\n- [ ] Verdict reached against the brief's stated success criteria\n- [ ] `poc-report.md` complete, with verbatim evidence and explicit non-coverage\n- [ ] `findings.md` updated during the work\n- [ ] Any skipped phase POC justified in `decisions.md` with a specific, checkable reason\n- [ ] Report presented at a gate and explicitly approved\n\n## Next\n\n- **Confirmed**, project-level → `\u002Fatom:plan`\n- **Confirmed**, phase-level → `\u002Fatom:spec`\n- **Refuted** → `\u002Fatom:discover`\n- **Inconclusive** → sharpen the question, respike\n","# \u002Fatom:poc\n\nThe **PROVE** beat. Builds a disposable Go program that answers exactly one question,\nruns it, and writes up what actually happened.\n\nUsed twice in the pipeline: once for the project, then once per phase that carries\ntechnical risk.\n\n## When to use\n\n- A POC Brief exists and has been approved\n- A phase in the plan has an open technical risk\n- Any moment someone is about to design around an unverified assumption\n\n## Writing artifacts\n\nSend complete POC reports, findings, decisions, and throwaway spike documentation as\nUTF-8 standard input to `atom-write-artifact \u003Crelative-path>`. Treat any refusal or\nnon-zero exit as a blocker; never bypass it with a generic editor or another shell\ncommand.\n\n## Before you start\n\nRead, in this order:\n\n1. [poc-discipline.md](..\u002Fatom\u002Freferences\u002Fpoc-discipline.md) — **the authority for this skill**\n2. [anti-rationalization.md](..\u002Fatom\u002Freferences\u002Fanti-rationalization.md)\n3. [verification.md](..\u002Fatom\u002Freferences\u002Fverification.md)\n4. [treatment-evidence.md](..\u002Fatom\u002Freferences\u002Ftreatment-evidence.md) — **required if the spike compares arms**\n\nYou need an approved brief. If there isn't one, go back to `\u002Fatom:discover`. Building a\nspike without an agreed question produces a program nobody can grade.\n\n---\n\n## Step 1 — Restate the question\n\nCopy the question out of the brief **verbatim** into your working notes.\n\nDo not reword it. Rewording is how a miss becomes a hit: the question drifts toward\nwhatever the spike turned out to be good at, and the report confirms something nobody\nasked.\n\n## Step 2 — Scaffold\n\n```bash\natom poc new {slug}\n```\n\nCreates `poc\u002F{slug}\u002F` with its own `go.mod`, a `main.go` stub, and a `README.md`\ncarrying the question. The spike is a **separate module** — it must not compile into\nthe production build, and it must be free to pull dependencies that production will\nnever take.\n\n## Step 3 — Write the spike\n\nWrite the smallest program that can answer the question.\n\n**This code is disposable, and should look it.** Hardcode values. Skip error handling.\n`panic` on anything unexpected. Put it all in `main.go`. Time spent making a spike\nelegant is time spent on code you are about to delete.\n\nThe one thing a spike must do well is **report**. It should print what it found,\nclearly enough that the output alone settles the question:\n\n```go\n\u002F\u002F The spike's entire job is to answer one question, so it prints the measurement\n\u002F\u002F the question asks about — not a generic \"done\".\nfmt.Printf(\"processed=%d duration=%s p99=%s\\n\", processed, elapsed, p99)\n```\n\nIf you find yourself building abstractions, splitting files, or adding configuration,\nthe question is too broad. Stop, narrow it, and check the narrowing with the developer.\n\n### The scope trap\n\nSpikes grow. You will be tempted to also check the adjacent thing while you're in\nhere. Don't. A spike that answers three questions answers none of them cleanly, and\nits report cannot be graded. Note the adjacent question in `findings.md` as a\ncandidate for a later spike.\n\n## Step 4 — Run it\n\n```bash\natom poc run {slug}\n```\n\nRuns `go vet`, `go build`, then `go run`, and captures **all** output.\n\n> **A POC that has not been executed does not exist.**\n\nReading the spike and concluding it would work is precisely the failure this whole\nframework is built to prevent. If it doesn't run, it has proven nothing.\n\nCapture the real output. You will paste it verbatim — never retyped, never rounded,\nnever reconstructed from memory. If you lost it, run it again.\n\nIf it measures something, run it more than once. A single timing sample is noise.\n\n### Comparison spikes: prove the treatment was applied\n\nSkip this if the spike has one arm. It is mandatory the moment the spike compares\narms and intends to credit the difference to a treatment.\n\nWork through the three checks in\n[treatment-evidence.md](..\u002Fatom\u002Freferences\u002Ftreatment-evidence.md) **before reading any\nresult**:\n\n| # | Check | How |\n|---|---|---|\n| 1 | The treatment reached the model | Render each arm's prompt, compare sizes |\n| 2 | Any required action actually happened | Count entries in the treatment's own usage log |\n| 3 | The treated arm behaved differently at all | Compare a behavioural measure against control |\n\nA treatment carried in the prompt arrives by construction and check 1 settles it. A\ntreatment the agent has to **do something** to use — run a script, query an index,\ncall an API — can fail silently, completing the run with no error while measuring\nnothing. Those treatments must log their own use.\n\n> **The agent's account of what it did is not evidence.** An agent that was blocked\n> may describe the action in the past tense and produce output in the right shape for\n> a tool it never ran.\n\nIf a check fails, the arm has proven nothing. Fix the cause, record the correction as\na deviation, and rerun — do not reinterpret the numbers you already have.\n\n## Step 5 — Reach a verdict\n\nThree outcomes. **All three are successes.**\n\n| Verdict | Meaning | Next |\n|---|---|---|\n| **Confirmed** | The assumption held, and the output shows it | Proceed to `\u002Fatom:spec` |\n| **Refuted** | The assumption was wrong | Back to `\u002Fatom:discover` with what you learned |\n| **Inconclusive** | The spike didn't isolate the question | Sharpen and respike. Do **not** proceed |\n\nGrade against the brief's success criteria, not against how you feel about the code.\n\n**Inconclusive is not a soft confirm.** If the spike crashed, or the numbers came out\nambiguous, or you only got it working by changing what you were testing — the verdict\nis Inconclusive. Writing that up as Confirmed launders a guess into the design as\nevidence, and every decision downstream inherits it.\n\n> Refuting your own plan is the highest-value outcome in this pipeline. You just saved\n> the design. Report it with enthusiasm, not apology.\n\n## Step 6 — Write the POC Report\n\nFill in [poc-report.md](..\u002Fatom\u002Ftemplates\u002Fpoc-report.md) at\n`specs\u002F{project}\u002Fpoc-report.md`, or `specs\u002F{project}\u002Fphase-NN-{slug}\u002Fpoc-report.md`\nfor a phase spike.\n\nTwo sections carry most of the value:\n\n- **Evidence** — verbatim output, in fenced blocks.\n- **Explicitly Not Covered** — the boundary of what this proves. A spike showing a\n  happy path at 1k req\u002Fs is not evidence about error paths at 100k req\u002Fs, and the\n  report must say so, or somebody will later believe it was.\n\nRecord anything that surprised you. Surprises are the highest-value content in the\nreport — they are, by definition, the things the design would otherwise have got wrong.\n\nAppend to `findings.md` **as you go**, not from memory afterwards.\n\n## Step 7 — Gate\n\nPresent per [human-gates.md](..\u002Fatom\u002Freferences\u002Fhuman-gates.md): the verdict, the\nevidence, what it changes, and one specific question.\n\n**End your turn. Wait.**\n\n---\n\n## After approval\n\nLeave the spike in `poc\u002F{slug}\u002F`. It is documentation now — the next person who\ndoubts the decision can re-run it.\n\n**Do not copy spike code into the implementation.** It earned its shortcuts by being\ndisposable. Production code earns its shape from the spec and the tests. Carry the\nknowledge forward, not the source.\n\n---\n\n## Done when\n\n- [ ] Question restated verbatim from the brief\n- [ ] Spike scaffolded as its own module under `poc\u002F{slug}\u002F`\n- [ ] Spike **executed**, with real output captured\n- [ ] For comparison spikes: treatment evidence collected for every arm before results were read\n- [ ] Verdict reached against the brief's stated success criteria\n- [ ] `poc-report.md` complete, with verbatim evidence and explicit non-coverage\n- [ ] `findings.md` updated during the work\n- [ ] Any skipped phase POC justified in `decisions.md` with a specific, checkable reason\n- [ ] Report presented at a gate and explicitly approved\n\n## Next\n\n- **Confirmed**, project-level → `\u002Fatom:plan`\n- **Confirmed**, phase-level → `\u002Fatom:spec`\n- **Refuted** → `\u002Fatom:discover`\n- **Inconclusive** → sharpen the question, respike\n",{"name":5,"description":7},"main","5bb33cbe2c835382b5ccda840c5268321b24ac0b","tools\u002Fatom\u002Fassets\u002Fskills\u002Fpoc","https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom","https:\u002F\u002Fgithub.com\u002Fgirishmotwani","curious about all things distributed ","2026-11-13 16:16:12",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add girishmotwani\u002Fatom --skill poc",{"tool":36,"command":37},"bunx","bunx skills add girishmotwani\u002Fatom --skill poc",{"tool":39,"command":40},"pnpm","pnpm dlx skills add girishmotwani\u002Fatom --skill poc","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom\u002Fblob\u002Fmain\u002Ftools\u002Fatom\u002Fassets\u002Fskills\u002Fpoc\u002FSKILL.md",true,[],[45,52,59,66,73,80],{"skillNo":46,"name":47,"nameZh":6,"description":48,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":49,"categories":50,"keywords":51,"publishedAt":18},"01M7DDBYEX4YKKBSQQP3HEGJC1","tdd","Implement one phase test-first in Go: red, green, refactor, against an approved design spec. Keywords: TDD, test driven, test first, implement, red green refactor, build the phase, write the code.","https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom\u002Fblob\u002Fmain\u002Ftools\u002Fatom\u002Fassets\u002Fskills\u002Ftdd\u002FSKILL.md",[],[],{"skillNo":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":18},"01M7DDBT5MMFRRBZJQ01VHNDGK","spec","Write the high-level design spec for one phase, grounded in that phase POC evidence. Keywords: spec, design, design doc, architecture, API design, describe the approach.","https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom\u002Fblob\u002Fmain\u002Ftools\u002Fatom\u002Fassets\u002Fskills\u002Fspec\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":18},"01M7DDBQ3NXQKG74B1RCHDKK9T","ship","Commit, push and open a pull request for a completed phase, with the pipeline artifacts linked. Keywords: ship, PR, pull request, commit, push, open a PR, submit.","https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom\u002Fblob\u002Fmain\u002Ftools\u002Fatom\u002Fassets\u002Fskills\u002Fship\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":18},"01M7DDBJMHEDT9VPGR1GJZD8CN","review","Standalone or phase multi-model review of a Go diff, merged into one report with severities. Keywords: review, code review, multi-model review, review my changes, review the diff, critique.","https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom\u002Fblob\u002Fmain\u002Ftools\u002Fatom\u002Fassets\u002Fskills\u002Freview\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":18},"01M7DDB98QNSK2WG5MMHH7S6J4","plan","Break proven work into ordered, independently shippable phases based on POC evidence. Keywords: plan, phases, break down, roadmap, decompose, milestones, sequencing.","https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom\u002Fblob\u002Fmain\u002Ftools\u002Fatom\u002Fassets\u002Fskills\u002Fplan\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":18},"01M7DDB3N1VYV1JK1K4Y2V3CP3","init","Onboard a Go repository to Atom: capture its initial baseline or confirm existing setup without changes. Keywords: init, onboard, setup, install atom, bootstrap, add atom to this repo.","https:\u002F\u002Fgithub.com\u002Fgirishmotwani\u002Fatom\u002Fblob\u002Fmain\u002Ftools\u002Fatom\u002Fassets\u002Fskills\u002Finit\u002FSKILL.md",[],[]]