[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79VTGCT56RFV4THB6RCEBJV":3,"skill-same:01M79VTGCT56RFV4THB6RCEBJV":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":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":6,"updatedAt":32,"install":33,"resources":46},"01M79VTGCT56RFV4THB6RCEBJV","define-loop",null,"Define a Loop — a durable scaffold that works a queue of recurring tasks autonomously (Sentry issues, Front tickets, small perf PRs) with its outputs held for human review. Use when someone wants an agent to keep doing a kind of work on a schedule, rather than one-off. Walks the interview, runs one real item live before arming anything, then writes the playbook and creates the loop.","yc-software\u002Fqm","yc-software","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F153323858?v=4","01M79VRJS45RJM7XBFGB9D1HY6","Y Combinator","01M79VRJSY4P8QKGTCMVJAZ4DE",15703,"MIT","https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Fdefine-loop\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: define-loop\ndescription: Define a Loop — a durable scaffold that works a queue of recurring tasks autonomously (Sentry issues, Front tickets, small perf PRs) with its outputs held for human review. Use when someone wants an agent to keep doing a kind of work on a schedule, rather than one-off. Walks the interview, runs one real item live before arming anything, then writes the playbook and creates the loop.\n---\n\n# define-loop — author a Loop\n\nA Loop is a cron with a memory and a gate: a trigger, a playbook, a ledger of work items\nthat survives across fires, ship-ready outputs held for a person, and a governor watching\nits health. Reach for one when the ask is \"keep doing this kind of work\", not \"do this\".\n\nIntake, work, and judge turns use the loop's existing owner\u002Fscope authorization,\ncredentials, command policy, and security posture. They run without surface tools or\naddressed delivery. Intake reads candidates; work prepares outputs; the judge inspects\nthe work and executes the declared checks. Final external actions belong in the\napproved ship stage. The review gate controls loop output shipping; it is not a\nread-only sandbox for arbitrary shell commands or API calls in earlier stages.\n\nA cron is enough when each fire is independent and nothing needs reviewing. Prefer the\ncron; a loop earns its complexity only when work items persist, must not be worked twice,\nand produce artifacts someone should see before they go out.\n\n**The one rule: never arm a loop you have not watched work.** Stage 2 is not optional.\n\n## Stage 1 — Interview\n\nGet these, in the user's own words. Ask about them one or two at a time, not as a form.\n\n| What                  | Why it matters                                                                                                                                                        |\n| --------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| **Source of work**    | How to enumerate candidates each fire, and the stable id per item (Sentry issue id, Front conversation id). That id is the dedupe key — get it right or work repeats. |\n| **Trigger**           | A cadence (calendar cron in their timezone) or an event. Ask when output is actually useful, not just how often.                                                      |\n| **Success condition** | What \"done\" means for ONE item.                                                                                                                                       |\n| **Ship actions**      | Which externally-visible actions the loop may take (`open_pr`, `send_email`, `front_reply`). Anything not declared is a quarantine when it happens.                   |\n| **Caps**              | Optional. Propose defaults; record a decline without argument.                                                                                                        |\n| **Escalation**        | Who the governor pings, and where.                                                                                                                                    |\n\n### Writing the success condition\n\nFree text, and it's the most load-bearing sentence in the loop. A good one has:\n\n- **one measurable end state** — a test result, a linked PR, an empty queue\n- **a stated check** — how the agent proves it (\"`npm test` exits 0\", \"the PR shows CI green\")\n- **the constraints that matter** — what must not change on the way (\"touches only the implicated module\")\n- **a bound** — \"or park with a diagnosis after 5 turns\", so a hopeless item stops burning money\n\n> The Sentry issue has a linked PR whose tests pass and CI is green, the fix touches only\n> the implicated module, or park with a diagnosis after 5 turns.\n\nAdd `successChecks` for anything a command can settle (`npm test`, `npm run lint`). The\njudge runs each check and reports its command and result. A missing or failed reported\ncheck prevents completion even when the judge claims success. Check results are\nmodel-reported, not an independent deterministic execution attestation.\n\n## Stage 2 — Shadow run (never skip)\n\nWork **one real item end to end, in this conversation, with the user watching.** No loop\nrecord exists yet.\n\n1. Pull one real candidate from the source.\n2. Do the whole job to the ship line — write the actual PR or the actual reply — and stop.\n3. Show the user the finished artifact. Ask directly: would you have sent this?\n4. Fix what they flag. Repeat on a second item if the first needed real correction.\n\nThen write the playbook **from what you just did** — the steps that actually worked, the\ndead ends worth skipping, the exact success condition you'd have wanted. A playbook\nwritten before the shadow run is a guess; one written after is a procedure.\n\n## Stage 3 — Emit\n\nCreate the loop with the `loop` tool: name, purpose, playbook, success condition, checks,\nship actions, caps, trigger. Every ship action starts at `hold` no matter what was asked\nfor — `auto` is earned in stage 4, never granted at creation.\n\nState back to the user, in one message: what it will do, when it fires, what it will\nproduce, what it will hold for them, and how to pause it.\n\n## Stage 4 — Supervised fires\n\nWatch the first few fires with the user.\n\n- Review the held outputs together the first time.\n- When a slice of work has been shipped unchanged several times running, offer graduation:\n  \"always ship `open_pr` labelled `lint-fix` from this loop\". That writes a standing grant\n  and flips only that slice to `auto`. If the org disables always-grants, graduation is\n  unavailable — say so plainly instead of half-doing it.\n- If outputs keep coming back returned, the playbook is wrong, not the reviewer. Fix the\n  playbook; a high return rate is the loudest signal a loop isn't working.\n\n## Editing a live loop\n\nPlaybook edits are versioned with provenance, from chat or the UI — same path, one\nhistory. \"Skip payments-service tickets from now on\" is a playbook edit, so make it one:\npropose the diff, get agreement, write it. Never let a loop carry behavior that isn't\nreadable in its playbook.\n\n## Refuse to arm\n\nSay no, and say why, when:\n\n- the user can't state a stable per-item id (work will repeat)\n- the success condition has no measurable end state (nothing can judge it)\n- `auto` is requested at creation on an externally-visible action (that's stage 4's job)\n- the shadow run never produced an artifact the user would actually send\n","# define-loop — author a Loop\n\nA Loop is a cron with a memory and a gate: a trigger, a playbook, a ledger of work items\nthat survives across fires, ship-ready outputs held for a person, and a governor watching\nits health. Reach for one when the ask is \"keep doing this kind of work\", not \"do this\".\n\nIntake, work, and judge turns use the loop's existing owner\u002Fscope authorization,\ncredentials, command policy, and security posture. They run without surface tools or\naddressed delivery. Intake reads candidates; work prepares outputs; the judge inspects\nthe work and executes the declared checks. Final external actions belong in the\napproved ship stage. The review gate controls loop output shipping; it is not a\nread-only sandbox for arbitrary shell commands or API calls in earlier stages.\n\nA cron is enough when each fire is independent and nothing needs reviewing. Prefer the\ncron; a loop earns its complexity only when work items persist, must not be worked twice,\nand produce artifacts someone should see before they go out.\n\n**The one rule: never arm a loop you have not watched work.** Stage 2 is not optional.\n\n## Stage 1 — Interview\n\nGet these, in the user's own words. Ask about them one or two at a time, not as a form.\n\n| What                  | Why it matters                                                                                                                                                        |\n| --------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| **Source of work**    | How to enumerate candidates each fire, and the stable id per item (Sentry issue id, Front conversation id). That id is the dedupe key — get it right or work repeats. |\n| **Trigger**           | A cadence (calendar cron in their timezone) or an event. Ask when output is actually useful, not just how often.                                                      |\n| **Success condition** | What \"done\" means for ONE item.                                                                                                                                       |\n| **Ship actions**      | Which externally-visible actions the loop may take (`open_pr`, `send_email`, `front_reply`). Anything not declared is a quarantine when it happens.                   |\n| **Caps**              | Optional. Propose defaults; record a decline without argument.                                                                                                        |\n| **Escalation**        | Who the governor pings, and where.                                                                                                                                    |\n\n### Writing the success condition\n\nFree text, and it's the most load-bearing sentence in the loop. A good one has:\n\n- **one measurable end state** — a test result, a linked PR, an empty queue\n- **a stated check** — how the agent proves it (\"`npm test` exits 0\", \"the PR shows CI green\")\n- **the constraints that matter** — what must not change on the way (\"touches only the implicated module\")\n- **a bound** — \"or park with a diagnosis after 5 turns\", so a hopeless item stops burning money\n\n> The Sentry issue has a linked PR whose tests pass and CI is green, the fix touches only\n> the implicated module, or park with a diagnosis after 5 turns.\n\nAdd `successChecks` for anything a command can settle (`npm test`, `npm run lint`). The\njudge runs each check and reports its command and result. A missing or failed reported\ncheck prevents completion even when the judge claims success. Check results are\nmodel-reported, not an independent deterministic execution attestation.\n\n## Stage 2 — Shadow run (never skip)\n\nWork **one real item end to end, in this conversation, with the user watching.** No loop\nrecord exists yet.\n\n1. Pull one real candidate from the source.\n2. Do the whole job to the ship line — write the actual PR or the actual reply — and stop.\n3. Show the user the finished artifact. Ask directly: would you have sent this?\n4. Fix what they flag. Repeat on a second item if the first needed real correction.\n\nThen write the playbook **from what you just did** — the steps that actually worked, the\ndead ends worth skipping, the exact success condition you'd have wanted. A playbook\nwritten before the shadow run is a guess; one written after is a procedure.\n\n## Stage 3 — Emit\n\nCreate the loop with the `loop` tool: name, purpose, playbook, success condition, checks,\nship actions, caps, trigger. Every ship action starts at `hold` no matter what was asked\nfor — `auto` is earned in stage 4, never granted at creation.\n\nState back to the user, in one message: what it will do, when it fires, what it will\nproduce, what it will hold for them, and how to pause it.\n\n## Stage 4 — Supervised fires\n\nWatch the first few fires with the user.\n\n- Review the held outputs together the first time.\n- When a slice of work has been shipped unchanged several times running, offer graduation:\n  \"always ship `open_pr` labelled `lint-fix` from this loop\". That writes a standing grant\n  and flips only that slice to `auto`. If the org disables always-grants, graduation is\n  unavailable — say so plainly instead of half-doing it.\n- If outputs keep coming back returned, the playbook is wrong, not the reviewer. Fix the\n  playbook; a high return rate is the loudest signal a loop isn't working.\n\n## Editing a live loop\n\nPlaybook edits are versioned with provenance, from chat or the UI — same path, one\nhistory. \"Skip payments-service tickets from now on\" is a playbook edit, so make it one:\npropose the diff, get agreement, write it. Never let a loop carry behavior that isn't\nreadable in its playbook.\n\n## Refuse to arm\n\nSay no, and say why, when:\n\n- the user can't state a stable per-item id (work will repeat)\n- the success condition has no measurable end state (nothing can judge it)\n- `auto` is requested at creation on an externally-visible action (that's stage 4's job)\n- the shadow run never produced an artifact the user would actually send\n",{"name":5,"description":7},"main","5011257b95913c04ebb800ac7bc80415546ee6bc","skills-seed\u002Fdefine-loop",1960,"https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm","https:\u002F\u002Flabs.ycombinator.com\u002Fqm","https:\u002F\u002Fgithub.com\u002Fyc-software","ycombinator.com","2026-11-15 01:58:49",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add yc-software\u002Fqm --skill define-loop",{"tool":39,"command":40},"bunx","bunx skills add yc-software\u002Fqm --skill define-loop",{"tool":42,"command":43},"pnpm","pnpm dlx skills add yc-software\u002Fqm --skill define-loop","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Fdefine-loop\u002FSKILL.md",true,[],[48,55,62,69,76,84],{"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},"01M79VXTEATDTE0X6SZRD4J75X","zoom","Find, create, reschedule, and cancel the user's Zoom meetings through their own Zoom connection.","https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Fzoom\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":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":19},"01M79VXKJ50M6G24Q11A3806R5","use-shared-credential","When you need to call a service the org has a SHARED credential for (a SERP\u002Fsearch key, a paid-API key, a data-vendor feed, a Git remote) — and the platform has told you that credential is available to this conversation — make the call BY PROXY through the credential broker. You never see the secret; the core stamps it onto the outbound request for you.","https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Fuse-shared-credential\u002FSKILL.md",[],[],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M79VXERKP52PBXTQ7KAX82S4","ui-state","See what the person is looking at in the QM web UI right now — the whole page (sidebar, every multiview pane, open dialogs) as raw HTML, the app's state as JSON, optionally the CSS and a screenshot. Use when they ask about something on their screen, want help with the UI, or you need to check what a canvas you wrote actually rendered.","https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Fui-state\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M79VXASHJ1WX4J68T3MBHZEK","ui-canvas","Draw your own HTML\u002FCSS\u002FJS as a full-window overlay on top of the QM web UI — a dashboard, a form, a visualization, controls that message you back. Write, update, replace, pin or dismiss it. Use when an interactive or visual result beats a chat reply.","https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Fui-canvas\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":80,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M79VX53549EYHNJBHD3PCFSG","taste-skill","Design taste and process for anything a person will look at in a browser — landing page, dashboard, prototype, deck. The default design driver for a browsable artifact, with an anti-slop playbook for the visual language.","https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Ftaste-skill\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Ftaste-skill\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},"01M79VX1DEE1BZZBK66NNCD238","slack-drafts","Create Slack message drafts in the user's own composer, as the user, through their per-user Slack OAuth. Drafting only — the user reviews and sends from Slack.","https:\u002F\u002Fgithub.com\u002Fyc-software\u002Fqm\u002Fblob\u002Fmain\u002Fskills-seed\u002Fslack-drafts\u002FSKILL.md",[],[]]