[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EB3R8KH19VGVEQAPZHC0SV":3,"skill-same:01M7EB3R8KH19VGVEQAPZHC0SV":64},{"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},"01M7EB3R8KH19VGVEQAPZHC0SV","call-state-reconciler",null,"Work out what actually happened to a CALL-E call by reading the call task, its attempts and its event stream together, and say which field each conclusion came from. Use when a workflow acts on a call result, when a call will not settle, or when the platform's own signals disagree with each other.","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\u002Fcall-state-reconciler\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: call-state-reconciler\ndescription: Work out what actually happened to a CALL-E call by reading the call task, its attempts and its event stream together, and say which field each conclusion came from. Use when a workflow acts on a call result, when a call will not settle, or when the platform's own signals disagree with each other.\nlicense: MIT\n---\n\n# Call State Reconciler\n\nA finished call hands you a status word, a `task_completed` boolean and a\nconfidence score. Three answers jammed into a shape that looks like one, and\nthe one everybody branches on is `task_completed`.\n\nIt comes back `true` for a voicemail box. The recording picks up, the agent\nasks its question into the tone, and the call ends `completed` with a\nschema-valid result and a high score. Nothing in that payload is a lie. There\nis simply no field in it that says a person was ever on the line.\n\nThis skill covers reading a call properly: how it ended, whether the job got\ndone, and whether the answer came from anybody speaking. Three separate\nquestions, answered separately, each one carrying the field it was read from.\n\n## When to use it\n\nUse it before a call result moves anything in a system of record. Booking,\ndispatching, charging, closing a ticket, marking a lead dead.\n\nUse it when a call will not settle, or when you are looking at a batch of calls\nand need to know which ones need a person today.\n\nUse it when two signals disagree: the call task says one thing and its recipient\nsays another, or the attempt timing does not match the event stream.\n\nDo not use it on a call nobody acts on automatically. A call that produces a\nnote for a human to read has no branch to get wrong, and this is overhead.\n\n## The one field that changes everything\n\nThe Calls API carries no answered-by field. The calls guide says so directly and\ntells you to define the classification yourself in a per-recipient result\nschema:\n\n```json\n{\n  \"type\": \"object\",\n  \"required\": [\"answered_by\"],\n  \"properties\": {\n    \"answered_by\": {\n      \"type\": \"string\",\n      \"enum\": [\"human\", \"ivr\", \"voicemail\", \"unknown\"],\n      \"description\": \"Classify the final endpoint. If an IVR transfers the call to a person, use human.\"\n    }\n  },\n  \"additionalProperties\": false\n}\n```\n\nSend that and the ending stops being a guess. Without it the honest answer to\n\"did a person pick up\" is that nothing in the payload establishes one, which is\ntrue and not much use to anybody.\n\nSo the first thing this skill does with a call it cannot resolve is name the\nfield that would resolve it. A check that only ever says no gets switched off.\n\nNote what `answered_by` is not. It is the platform's classification of what it\nheard, so it can be wrong. What changes is who is doing the guessing: with the\nfield declared the claim is stated in the payload and can be pointed at, instead\nof being inferred by whatever read it.\n\n## The four things it reconciles\n\nEach one is a state somebody has reported and nothing has closed.\n\n**Nobody is watching it.** The call is still open, and long enough has passed\nthat whatever was waiting has already given up. A client timeout is not the same\nas the call ending, and a call can dial well after the SDK has raised.\n\n**It may have gone out twice.** The returned call is older than the request that\nreturned it, so the platform handed back one that already existed. A replay\notherwise reads exactly like a fresh creation.\n\n**The clock contradicts itself.** An attempt that claims to have run longer than\nthe call existed, or one timestamped outside the window the events describe.\nOnly contradictions count. An event stream wider than the attempt window is\nnormal, because the stream also covers queuing and finalization.\n\n**The retry is not safe.** The submission is in an unknown state, or the same\nidempotency key was refused after a failure. Deriving the key from the\nauthorization rather than the attempt is what makes the next attempt answerable:\nsame approved action, same key, so a retry cannot become a second call.\n\n## Read the event stream, not the attempt clock\n\n`GET \u002Fv1\u002Fcalls\u002F{call_id}\u002Fevents` carries a `created_at` on every event, declared\n`format: date-time`, and those timestamps hold still.\n\nThe attempt timestamps do not. On a finished call they can be rewritten shortly\nafter it settles: the offset is dropped, the value moves, and sub-second\nprecision is lost to rounding. Read the same call twice and you can get two\ndifferent answers about how long it ran. A sub-second attempt rounds to zero.\n\nSo reconstruct timing from events and treat the attempt fields as evidence about\nthe attempt rather than as a clock. `references\u002Fobserved-shapes.md` shows the\nshape with invented values and how to check it on your own account.\n\n## What it will not do\n\nIt reads fields. It does not interpret a transcript, score anything, or ask a\nmodel what it thinks, so every sentence it produces traces to a value in the\npayload and says the same thing every time.\n\nIt never resolves the business outcome. Where a reading rests on inference it is\nmarked as inference and routed to a person, and where no field carries the fact\nat all it says that rather than filling the gap.\n\nIt places no calls.\n\n## References\n\n- `references\u002Fobserved-shapes.md` - the platform behaviours this relies on, with\n  invented values, and how to check each one yourself.\n- `references\u002Freading-a-call.md` - the three axes, the endings each surface can\n  and cannot express, and worked examples.\n","# Call State Reconciler\n\nA finished call hands you a status word, a `task_completed` boolean and a\nconfidence score. Three answers jammed into a shape that looks like one, and\nthe one everybody branches on is `task_completed`.\n\nIt comes back `true` for a voicemail box. The recording picks up, the agent\nasks its question into the tone, and the call ends `completed` with a\nschema-valid result and a high score. Nothing in that payload is a lie. There\nis simply no field in it that says a person was ever on the line.\n\nThis skill covers reading a call properly: how it ended, whether the job got\ndone, and whether the answer came from anybody speaking. Three separate\nquestions, answered separately, each one carrying the field it was read from.\n\n## When to use it\n\nUse it before a call result moves anything in a system of record. Booking,\ndispatching, charging, closing a ticket, marking a lead dead.\n\nUse it when a call will not settle, or when you are looking at a batch of calls\nand need to know which ones need a person today.\n\nUse it when two signals disagree: the call task says one thing and its recipient\nsays another, or the attempt timing does not match the event stream.\n\nDo not use it on a call nobody acts on automatically. A call that produces a\nnote for a human to read has no branch to get wrong, and this is overhead.\n\n## The one field that changes everything\n\nThe Calls API carries no answered-by field. The calls guide says so directly and\ntells you to define the classification yourself in a per-recipient result\nschema:\n\n```json\n{\n  \"type\": \"object\",\n  \"required\": [\"answered_by\"],\n  \"properties\": {\n    \"answered_by\": {\n      \"type\": \"string\",\n      \"enum\": [\"human\", \"ivr\", \"voicemail\", \"unknown\"],\n      \"description\": \"Classify the final endpoint. If an IVR transfers the call to a person, use human.\"\n    }\n  },\n  \"additionalProperties\": false\n}\n```\n\nSend that and the ending stops being a guess. Without it the honest answer to\n\"did a person pick up\" is that nothing in the payload establishes one, which is\ntrue and not much use to anybody.\n\nSo the first thing this skill does with a call it cannot resolve is name the\nfield that would resolve it. A check that only ever says no gets switched off.\n\nNote what `answered_by` is not. It is the platform's classification of what it\nheard, so it can be wrong. What changes is who is doing the guessing: with the\nfield declared the claim is stated in the payload and can be pointed at, instead\nof being inferred by whatever read it.\n\n## The four things it reconciles\n\nEach one is a state somebody has reported and nothing has closed.\n\n**Nobody is watching it.** The call is still open, and long enough has passed\nthat whatever was waiting has already given up. A client timeout is not the same\nas the call ending, and a call can dial well after the SDK has raised.\n\n**It may have gone out twice.** The returned call is older than the request that\nreturned it, so the platform handed back one that already existed. A replay\notherwise reads exactly like a fresh creation.\n\n**The clock contradicts itself.** An attempt that claims to have run longer than\nthe call existed, or one timestamped outside the window the events describe.\nOnly contradictions count. An event stream wider than the attempt window is\nnormal, because the stream also covers queuing and finalization.\n\n**The retry is not safe.** The submission is in an unknown state, or the same\nidempotency key was refused after a failure. Deriving the key from the\nauthorization rather than the attempt is what makes the next attempt answerable:\nsame approved action, same key, so a retry cannot become a second call.\n\n## Read the event stream, not the attempt clock\n\n`GET \u002Fv1\u002Fcalls\u002F{call_id}\u002Fevents` carries a `created_at` on every event, declared\n`format: date-time`, and those timestamps hold still.\n\nThe attempt timestamps do not. On a finished call they can be rewritten shortly\nafter it settles: the offset is dropped, the value moves, and sub-second\nprecision is lost to rounding. Read the same call twice and you can get two\ndifferent answers about how long it ran. A sub-second attempt rounds to zero.\n\nSo reconstruct timing from events and treat the attempt fields as evidence about\nthe attempt rather than as a clock. `references\u002Fobserved-shapes.md` shows the\nshape with invented values and how to check it on your own account.\n\n## What it will not do\n\nIt reads fields. It does not interpret a transcript, score anything, or ask a\nmodel what it thinks, so every sentence it produces traces to a value in the\npayload and says the same thing every time.\n\nIt never resolves the business outcome. Where a reading rests on inference it is\nmarked as inference and routed to a person, and where no field carries the fact\nat all it says that rather than filling the gap.\n\nIt places no calls.\n\n## References\n\n- `references\u002Fobserved-shapes.md` - the platform behaviours this relies on, with\n  invented values, and how to check each one yourself.\n- `references\u002Freading-a-call.md` - the three axes, the endings each surface can\n  and cannot express, and worked examples.\n",{"name":5,"license":15,"description":7},"master","8c91ae30409640de28e218b3e9c7fe01c01d257b","skills\u002Fcall-state-reconciler","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 call-state-reconciler",{"tool":38,"command":39},"bunx","bunx skills add fozagtx\u002FBuildOS-workflow --skill call-state-reconciler",{"tool":41,"command":42},"pnpm","pnpm dlx skills add fozagtx\u002FBuildOS-workflow --skill call-state-reconciler","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fcall-state-reconciler\u002FSKILL.md",true,[46,52,56,60],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fexamples.md","REFERENCE","参考文档",3788,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fcall-state-reconciler\u002Freferences\u002Fexamples.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fobserved-shapes.md",3648,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fcall-state-reconciler\u002Freferences\u002Fobserved-shapes.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Freading-a-call.md",2884,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fcall-state-reconciler\u002Freferences\u002Freading-a-call.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Fsafety.md",3853,"https:\u002F\u002Fgithub.com\u002Ffozagtx\u002FBuildOS-workflow\u002Fblob\u002Fmaster\u002Fskills\u002Fcall-state-reconciler\u002Freferences\u002Fsafety.md",[65,72,79,86,93,102],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"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":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"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":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"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":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"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":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":97,"licenseFileUrl":98,"sourceHtmlUrl":99,"categories":100,"keywords":101,"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":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":108,"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",[],[]]