[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EC864ASPZVCEHZ2H1DC3RC":3,"skill-same:01M7EC864ASPZVCEHZ2H1DC3RC":72},{"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":29,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M7EC864ASPZVCEHZ2H1DC3RC","holdfor-post-visit-followup",null,"Place one consent-gated post-appointment check-in phone call to an older patient on behalf of a practice, return five enumerated answers plus a verbatim patient quote, and stop the call rather than answer anything clinical. Then, only after a named human releases it, place a second call into the practice's own booking line carrying that quote and a bounded range of dates. Never books unilaterally, never advises, never asks the patient to confirm personal details.","fozagtx\u002FBuildOS-workflow","fozagtx","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160292135?v=4","01M7EAS4W39D2X3WNCCVVZ7CVZ","fawuzan","01M7EAS4WP1G4JD4KG87P6PTK6",0,"MIT","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fholdfor-post-visit-followup\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: holdfor-post-visit-followup\ndescription: Place one consent-gated post-appointment check-in phone call to an older patient on behalf of a practice, return five enumerated answers plus a verbatim patient quote, and stop the call rather than answer anything clinical. Then, only after a named human releases it, place a second call into the practice's own booking line carrying that quote and a bounded range of dates. Never books unilaterally, never advises, never asks the patient to confirm personal details.\nlicense: MIT\n---\n\n# HoldFor post-visit follow-up\n\nTwo phone calls with a named human between them.\n\n1. The **Check-in Call** rings a patient a few days after an appointment, asks five\n   bounded questions at her pace, and returns a structured result.\n2. A member of practice staff reads it and grants a **Release**, or does not.\n3. The **Rebooking Call** rings the practice's own booking line inside that Release,\n   carrying her own words.\n\nNothing in step 3 can happen without step 2. That is the whole shape of the skill.\n\nTerms in bold with a capital are defined in [`CONTEXT.md`](..\u002F..\u002Fdocs\u002Fholdfor-post-visit-followup\u002FCONTEXT.md) and are\nload-bearing. Do not substitute \"ticket\" for **Review Item**, \"approval\" for\n**Release**, or \"escalation\" for **Stop Condition**.\n\n## What this skill does not do\n\nNeither call decides anything clinical, and neither writes to a patient record. There is\nno write path to the clinical record, by design and not by omission. See\n[ADR 0001](..\u002F..\u002Fdocs\u002Fadr\u002F0001-no-agent-write-path-to-the-clinical-record.md).\n\nThe Check-in Call carries no authority to cause the Rebooking Call. A patient saying yes\nis an answer, not an instruction — only a **Release**, granted by a named human who read\nthe transcript, places the second call. See\n[ADR 0003](..\u002F..\u002Fdocs\u002Fadr\u002F0003-a-release-grants-a-bounded-authority.md).\n\n## When to use\n\n- An appointment happened, the practice operates this workflow, and the patient has\n  recorded consent to be called.\n- It is the patient's **Due Day** — day 3 after the appointment, stepped forward off a\n  weekend. One day, and the call is not placed on any other.\n- The patient is an adult with an existing care relationship with the practice.\n\n## When not to use\n\n- No recorded consent. There is no fallback and no implied consent; the call is simply\n  not placed.\n- Any day but the Due Day, weekends included. This is a refusal, not an error.\n- The patient can use the online portal comfortably. Then \"use the portal\" is a\n  complete answer and this call has no reason to exist.\n- Anything clinical is being asked of the agent. Read the **Safety Line** and stop.\n- Cold outreach, marketing, screening a population, or any call to someone the\n  practice has no existing relationship with.\n\n## Call one — the Check-in Call\n\nRead [`references\u002Fcall-task.md`](references\u002Fcall-task.md) for the authored wording. It\nis the source of truth for what the agent says, and the sentences marked fixed are\npinned by tests because each one carries a promise.\n\nFive questions, four bounded and one open:\n\n| # | Question | Field |\n| --- | --- | --- |\n| 1 | Since {weekday}, are you feeling better, about the same, or worse? | `feeling` |\n| 2 | Are you getting on alright with what they gave you? | `medication_ok` |\n| 3 | Is there anything worrying you? | **Carried Words** |\n| 4 | Would you like the surgery to see you again? | `wants_seen` |\n| 5 | Are mornings or afternoons easier for you? | `when_easier` |\n\nTwo of them are conditional, and the condition is enforced after the call rather than\ntrusted to the agent. Question 2 is asked only when the appointment changed her\nmedication. Question 5 is asked only when she said yes to question 4 — putting it to\nsomebody who has just declined presses her towards an appointment she did not want.\nEither way the field records `not_asked`, which means *nobody put this to her*, not\n*she had no answer*.\n\nQuestion 5 books nothing and promises nothing. The agent may not say a time is\navailable, offer one, or tell her when she will be seen: no human has read the call yet.\nA patient who names a weekday rather than a half of the day is recorded `unsure` — the\nenvelope has no room for a weekday, and inventing a half-day would put a constraint in\nher mouth she never gave.\n\nThe result shape is [`references\u002Fresult-schema.json`](references\u002Fresult-schema.json).\n\n## Call two — the Rebooking Call\n\nRead [`references\u002Fenvelope.md`](references\u002Fenvelope.md) for the contract: what a Release\ncarries, how an offered slot is matched, and what the outcome may be.\n\nThree rules stand out and are worth knowing before reading it.\n\n**It speaks from a closed list.** The practice's own name, what the Release contains,\nand the patient's identifiers. Nothing else. Anything reception asks outside those three\nis answered with one fixed sentence and then the call returns to scheduling. The name\ngoes out in the opening; the date of birth waits to be asked, so a call that reached a\nwrong number has disclosed a name and nothing more. See\n[ADR 0011](..\u002F..\u002Fdocs\u002Fadr\u002F0011-the-rebooking-call-speaks-from-a-closed-list.md).\n\n**An offered slot is matched, never resolved.** \"Wednesday the 26th\" is not a date. The\nagent does not turn it into one — the envelope is asked whether it holds a day this turn\nnamed, and two candidates means neither. See\n[ADR 0008](..\u002F..\u002Fdocs\u002Fadr\u002F0008-an-offered-slot-is-matched-never-resolved.md).\n\n**Reception revises, so every offer is kept.** The Binding Acceptance is the *last*\naccepted offer; the earlier ones stay as evidence. When a human rings back, the\nwithdrawn 09:10 is the reason the booking is 08:50. See\n[ADR 0012](..\u002F..\u002Fdocs\u002Fadr\u002F0012-the-last-acceptance-binds-and-every-offer-is-kept.md).\n\n## The Never-Ask Rule\n\nThe check-in call asks for no date of birth, no address, no NHS number, and nothing\nresembling payment — and it says so aloud.\n\nThis inverts the usual \"verify the caller first\" instinct deliberately. An unexpected\nvoice asking an older person to confirm personal details is the most recognisable\nopening of a telephone scam there is, so a genuine call that opens that way cannot be\ntold apart from a fraudulent one. **The call that cannot ask is the only one that can\nbe trusted.** The agent proves itself instead with a fact only the practice holds: the\nday she was seen.\n\nIdentifiers that reception needs are supplied later by the Rebooking Call, from the\npractice's own record. They are never obtained from the patient.\n\n## Carried Words are verbatim or nothing\n\nThe answer to question 3 is stored as a substring of a single patient turn, together\nwith that turn's index — or not stored at all. Never generated, never summarised,\nnever tidied into better English.\n\nNo span is a valid outcome. The Review Item then queues for a human instead of\ncarrying an invented quote forward. This is not fussiness: the stored string is later\nspoken aloud to a receptionist in the patient's name, so a quote she never said would\nput words in her mouth to a third party.\n\nA Reviewer may narrow the quote before releasing it, or release none of it at all. See\n[ADR 0009](..\u002F..\u002Fdocs\u002Fadr\u002F0009-the-board-defaults-to-silence.md).\n\n## Stop Conditions are enforced twice\n\nThe two layers are not interchangeable, and neither replaces the other. See\n[ADR 0005](..\u002F..\u002Fdocs\u002Fadr\u002F0005-stop-conditions-are-enforced-twice.md).\n\n1. **In the call, for the patient.** The agent reads the fixed Safety Line and ends\n   the call kindly. Nothing is improvised — a model improvising here is a model giving\n   medical advice.\n2. **After the call, for the practice.** A deterministic scanner reads the finished\n   transcript and decides whether the Review Item is flagged. It is the authority, and\n   it flags even when the agent sailed through all five questions without stopping.\n\nA Stop Condition is a surface condition, not a judgement: a phrase on the red-flag\nlist, an answer that maps to no bounded field, repeated confusion, a third party on the\nline, or a clinical question put to the agent. The agent matches. It never grades\nseverity.\n\nThe Safety Line is placed **before** the red-flag list in the prompt, not after it. An\nearly \"stop the call\" imperative followed by ninety-odd phrases is a model that stops\nwithout speaking, and a line going dead is the worst answer available to somebody who\nhas just described chest pain. The line names where to turn, ends with her name the way\nevery other exit from the call does, and is said in full before the call ends.\n\nA flagged Review Item cannot be released at all. The board's answer to it is \"ring them\nmyself\", not a second agent placed against the call the first one correctly refused.\n\nUse [`references\u002Fred-flags.md`](references\u002Fred-flags.md) as the single list. Both the\nprompt builder and the scanner read it through the same loader, so the two cannot\ndrift apart. Do not hand-copy phrases into a prompt.\n\n## Safety\n\nRead [`references\u002Fsafety.md`](references\u002Fsafety.md) before placing any live call.\nWorked results, including the outcomes that look like failures and are not, are in\n[`references\u002Fexamples.md`](references\u002Fexamples.md).\n","# HoldFor post-visit follow-up\n\nTwo phone calls with a named human between them.\n\n1. The **Check-in Call** rings a patient a few days after an appointment, asks five\n   bounded questions at her pace, and returns a structured result.\n2. A member of practice staff reads it and grants a **Release**, or does not.\n3. The **Rebooking Call** rings the practice's own booking line inside that Release,\n   carrying her own words.\n\nNothing in step 3 can happen without step 2. That is the whole shape of the skill.\n\nTerms in bold with a capital are defined in [`CONTEXT.md`](..\u002F..\u002Fdocs\u002Fholdfor-post-visit-followup\u002FCONTEXT.md) and are\nload-bearing. Do not substitute \"ticket\" for **Review Item**, \"approval\" for\n**Release**, or \"escalation\" for **Stop Condition**.\n\n## What this skill does not do\n\nNeither call decides anything clinical, and neither writes to a patient record. There is\nno write path to the clinical record, by design and not by omission. See\n[ADR 0001](..\u002F..\u002Fdocs\u002Fadr\u002F0001-no-agent-write-path-to-the-clinical-record.md).\n\nThe Check-in Call carries no authority to cause the Rebooking Call. A patient saying yes\nis an answer, not an instruction — only a **Release**, granted by a named human who read\nthe transcript, places the second call. See\n[ADR 0003](..\u002F..\u002Fdocs\u002Fadr\u002F0003-a-release-grants-a-bounded-authority.md).\n\n## When to use\n\n- An appointment happened, the practice operates this workflow, and the patient has\n  recorded consent to be called.\n- It is the patient's **Due Day** — day 3 after the appointment, stepped forward off a\n  weekend. One day, and the call is not placed on any other.\n- The patient is an adult with an existing care relationship with the practice.\n\n## When not to use\n\n- No recorded consent. There is no fallback and no implied consent; the call is simply\n  not placed.\n- Any day but the Due Day, weekends included. This is a refusal, not an error.\n- The patient can use the online portal comfortably. Then \"use the portal\" is a\n  complete answer and this call has no reason to exist.\n- Anything clinical is being asked of the agent. Read the **Safety Line** and stop.\n- Cold outreach, marketing, screening a population, or any call to someone the\n  practice has no existing relationship with.\n\n## Call one — the Check-in Call\n\nRead [`references\u002Fcall-task.md`](references\u002Fcall-task.md) for the authored wording. It\nis the source of truth for what the agent says, and the sentences marked fixed are\npinned by tests because each one carries a promise.\n\nFive questions, four bounded and one open:\n\n| # | Question | Field |\n| --- | --- | --- |\n| 1 | Since {weekday}, are you feeling better, about the same, or worse? | `feeling` |\n| 2 | Are you getting on alright with what they gave you? | `medication_ok` |\n| 3 | Is there anything worrying you? | **Carried Words** |\n| 4 | Would you like the surgery to see you again? | `wants_seen` |\n| 5 | Are mornings or afternoons easier for you? | `when_easier` |\n\nTwo of them are conditional, and the condition is enforced after the call rather than\ntrusted to the agent. Question 2 is asked only when the appointment changed her\nmedication. Question 5 is asked only when she said yes to question 4 — putting it to\nsomebody who has just declined presses her towards an appointment she did not want.\nEither way the field records `not_asked`, which means *nobody put this to her*, not\n*she had no answer*.\n\nQuestion 5 books nothing and promises nothing. The agent may not say a time is\navailable, offer one, or tell her when she will be seen: no human has read the call yet.\nA patient who names a weekday rather than a half of the day is recorded `unsure` — the\nenvelope has no room for a weekday, and inventing a half-day would put a constraint in\nher mouth she never gave.\n\nThe result shape is [`references\u002Fresult-schema.json`](references\u002Fresult-schema.json).\n\n## Call two — the Rebooking Call\n\nRead [`references\u002Fenvelope.md`](references\u002Fenvelope.md) for the contract: what a Release\ncarries, how an offered slot is matched, and what the outcome may be.\n\nThree rules stand out and are worth knowing before reading it.\n\n**It speaks from a closed list.** The practice's own name, what the Release contains,\nand the patient's identifiers. Nothing else. Anything reception asks outside those three\nis answered with one fixed sentence and then the call returns to scheduling. The name\ngoes out in the opening; the date of birth waits to be asked, so a call that reached a\nwrong number has disclosed a name and nothing more. See\n[ADR 0011](..\u002F..\u002Fdocs\u002Fadr\u002F0011-the-rebooking-call-speaks-from-a-closed-list.md).\n\n**An offered slot is matched, never resolved.** \"Wednesday the 26th\" is not a date. The\nagent does not turn it into one — the envelope is asked whether it holds a day this turn\nnamed, and two candidates means neither. See\n[ADR 0008](..\u002F..\u002Fdocs\u002Fadr\u002F0008-an-offered-slot-is-matched-never-resolved.md).\n\n**Reception revises, so every offer is kept.** The Binding Acceptance is the *last*\naccepted offer; the earlier ones stay as evidence. When a human rings back, the\nwithdrawn 09:10 is the reason the booking is 08:50. See\n[ADR 0012](..\u002F..\u002Fdocs\u002Fadr\u002F0012-the-last-acceptance-binds-and-every-offer-is-kept.md).\n\n## The Never-Ask Rule\n\nThe check-in call asks for no date of birth, no address, no NHS number, and nothing\nresembling payment — and it says so aloud.\n\nThis inverts the usual \"verify the caller first\" instinct deliberately. An unexpected\nvoice asking an older person to confirm personal details is the most recognisable\nopening of a telephone scam there is, so a genuine call that opens that way cannot be\ntold apart from a fraudulent one. **The call that cannot ask is the only one that can\nbe trusted.** The agent proves itself instead with a fact only the practice holds: the\nday she was seen.\n\nIdentifiers that reception needs are supplied later by the Rebooking Call, from the\npractice's own record. They are never obtained from the patient.\n\n## Carried Words are verbatim or nothing\n\nThe answer to question 3 is stored as a substring of a single patient turn, together\nwith that turn's index — or not stored at all. Never generated, never summarised,\nnever tidied into better English.\n\nNo span is a valid outcome. The Review Item then queues for a human instead of\ncarrying an invented quote forward. This is not fussiness: the stored string is later\nspoken aloud to a receptionist in the patient's name, so a quote she never said would\nput words in her mouth to a third party.\n\nA Reviewer may narrow the quote before releasing it, or release none of it at all. See\n[ADR 0009](..\u002F..\u002Fdocs\u002Fadr\u002F0009-the-board-defaults-to-silence.md).\n\n## Stop Conditions are enforced twice\n\nThe two layers are not interchangeable, and neither replaces the other. See\n[ADR 0005](..\u002F..\u002Fdocs\u002Fadr\u002F0005-stop-conditions-are-enforced-twice.md).\n\n1. **In the call, for the patient.** The agent reads the fixed Safety Line and ends\n   the call kindly. Nothing is improvised — a model improvising here is a model giving\n   medical advice.\n2. **After the call, for the practice.** A deterministic scanner reads the finished\n   transcript and decides whether the Review Item is flagged. It is the authority, and\n   it flags even when the agent sailed through all five questions without stopping.\n\nA Stop Condition is a surface condition, not a judgement: a phrase on the red-flag\nlist, an answer that maps to no bounded field, repeated confusion, a third party on the\nline, or a clinical question put to the agent. The agent matches. It never grades\nseverity.\n\nThe Safety Line is placed **before** the red-flag list in the prompt, not after it. An\nearly \"stop the call\" imperative followed by ninety-odd phrases is a model that stops\nwithout speaking, and a line going dead is the worst answer available to somebody who\nhas just described chest pain. The line names where to turn, ends with her name the way\nevery other exit from the call does, and is said in full before the call ends.\n\nA flagged Review Item cannot be released at all. The board's answer to it is \"ring them\nmyself\", not a second agent placed against the call the first one correctly refused.\n\nUse [`references\u002Fred-flags.md`](references\u002Fred-flags.md) as the single list. Both the\nprompt builder and the scanner read it through the same loader, so the two cannot\ndrift apart. Do not hand-copy phrases into a prompt.\n\n## Safety\n\nRead [`references\u002Fsafety.md`](references\u002Fsafety.md) before placing any live call.\nWorked results, including the outcomes that look like failures and are not, are in\n[`references\u002Fexamples.md`](references\u002Fexamples.md).\n",{"name":5,"license":15,"description":7},"master","8c91ae30409640de28e218b3e9c7fe01c01d257b","skills\u002Fholdfor-post-visit-followup","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow","https:\u002F\u002Fgithub.com\u002Ffozagtx","hi.fawuzanpima@gmail.com","Forward deployed","2026-10-08 00:13:28",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add fozagtx\u002FBuildOS-workflow --skill holdfor-post-visit-followup",{"tool":38,"command":39},"bunx","bunx skills add fozagtx\u002FBuildOS-workflow --skill holdfor-post-visit-followup",{"tool":41,"command":42},"pnpm","pnpm dlx skills add fozagtx\u002FBuildOS-workflow --skill holdfor-post-visit-followup","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fholdfor-post-visit-followup\u002FSKILL.md",true,[46,52,56,60,64,68],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fcall-task.md","REFERENCE","参考文档",12474,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fholdfor-post-visit-followup\u002Freferences\u002Fcall-task.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fenvelope.md",7566,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fholdfor-post-visit-followup\u002Freferences\u002Fenvelope.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fexamples.md",5098,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fholdfor-post-visit-followup\u002Freferences\u002Fexamples.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Fred-flags.md",7196,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fholdfor-post-visit-followup\u002Freferences\u002Fred-flags.md",{"path":65,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":66,"sourceHtmlUrl":67},"references\u002Fresult-schema.json",4275,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fholdfor-post-visit-followup\u002Freferences\u002Fresult-schema.json",{"path":69,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":70,"sourceHtmlUrl":71},"references\u002Fsafety.md",5854,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fholdfor-post-visit-followup\u002Freferences\u002Fsafety.md",[73,80,87,94,101,110],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7EE49XAN8GE213J691P3RZ9","xitter","Interact with X\u002FTwitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks, mentions, and user lookups.","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fxitter\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7EE3TMR3CR7HR7HTCAKPTTR","writing-plans","Use when you have a spec or requirements for a multi-step task. Creates comprehensive implementation plans with bite-sized tasks, exact file paths, and complete code examples.","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fwriting-plans\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M7EE3AJ6V2VBVAX1WTV7M447","whisper","OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fwhisper\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":19},"01M7EE36AAP5THSCQAHH6JCWVV","weights-and-biases","Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fweights-and-biases\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":105,"licenseFileUrl":106,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":19},"01M7EE30R3S4EAK1FWJEBG7MW9","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fwebapp-testing\u002FSKILL.md",[],[],{"skillNo":111,"name":112,"nameZh":6,"description":113,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":114,"categories":115,"keywords":116,"publishedAt":19},"01M7EE2X3PJ0ZJR0JJNF2M4NS5","web-pentest","Authorized web pentest: recon, proof-based exploits, report.","https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fweb-pentest\u002FSKILL.md",[],[]]