[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71AYB9CV388KZ1JCBA2TYVZ":3,"skill-same:01M71AYB9CV388KZ1JCBA2TYVZ":85},{"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":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":61,"commitSha":62,"skillDirectory":63,"forksCount":64,"repositoryHtmlUrl":65,"repositoryHomepageUrl":66,"ownerHtmlUrl":67,"ownerBlogUrl":68,"ownerDescription":69,"updatedAt":70,"install":71,"resources":84},"01M71AYB9CV388KZ1JCBA2TYVZ","quota-resilience",null,"Salvage and resume when the session hits its own usage limit \u002F quota hard\nstop: commit complete work, leave a clean tree, write a committed resume\nnote, report exact state, and offer one bounded auto-restart where the host\ncan schedule one. Use when the agent's own usage limit, quota, or rate\nbudget is exhausted mid-task, or when picking up work a session left when\nit died that way. Not for HTTP 429s \u002F rate limits the user's application\nreceives — that is application behavior (\u002Fdebug, \u002Fqa), not agent state.","ariadoss\u002Fsuperskills","ariadoss","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F414341?v=4","01M71AC7383PT6JCRZSQJ4AVTM","Danilo Stern-Sapad","01M71AC73Y7NTBTJ9KQYNENRRE",12,"MIT","https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills\u002Fblob\u002Fmain\u002Fskills\u002Fquota-resilience\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000038","人力资源","PROFESSION","适用职业",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"quota","resilience","quota resilience","commit","Bash","when","debug","http","use","auto","host","qa","clean","up","session","complete","for","limit","usage","tree","2026-11-10 10:12:30",false,"---\nname: quota-resilience\nversion: 1.1.0\ndescription: |\n  Salvage and resume when the session hits its own usage limit \u002F quota hard\n  stop: commit complete work, leave a clean tree, write a committed resume\n  note, report exact state, and offer one bounded auto-restart where the host\n  can schedule one. Use when the agent's own usage limit, quota, or rate\n  budget is exhausted mid-task, or when picking up work a session left when\n  it died that way. Not for HTTP 429s \u002F rate limits the user's application\n  receives — that is application behavior (\u002Fdebug, \u002Fqa), not agent state.\ntriggers:\n  - hit my usage limit\n  - usage limit hit\n  - quota exhausted\n  - quota hard stop\n  - session died mid-task\n  - resume after the limit resets\nallowed-tools:\n  - Bash\n  - Read\n  - Write\n---\n\n# \u002Fquota-resilience\n\nA usage-limit hard stop is a shutdown, not an obstacle to route around.\nRetrying the blocked call, downgrading the approach, or improvising \"one\nmore push\" spends the remaining window and makes the resume expensive. This\nskill spends that window making the resume cheap instead: committed work,\nclean tree, a note that carries the state, and one bounded restart where the\nhost can schedule one.\n\n## Phase 1 — on the limit error (stop)\n\nHard rules, in order:\n\n1. **Stop initiating work.** No retries of the blocked call, no new\n   subagents, no reduced-effort substitutes for the blocked task. Every\n   further call either fails identically or steals the window the shutdown\n   needs.\n2. **Salvage.** Split the working tree:\n   - *Complete* — passes its own check (tests, lint, build) or stands alone\n     (docs, config): commit with an honest message.\n   - *Incomplete* — never left dangling. Commit as a clearly labeled `wip:`\n     checkpoint if it is a useful base, or revert it if it is noise. Either\n     way `git status` ends **clean**: the next session starts from committed\n     state, never from a diff it has to reverse-engineer.\n   - Do not start new work to \"round out\" a commit.\n3. **Write the resume note** — `QUOTA-RESUME.md` at the repo root, committed\n   with the salvage. If this session's `handoff-\u003Csession-id>.md` (or a `handoff-*.md` the\n   work clearly belongs to) exists (written by `\u002Fhandoff` before the\n   stop), fold its content into this note and delete that file — one bridge,\n   not two. Leave other sessions' handoff files alone. Required sections:\n   - **Goal** — one line: the task's definition of done.\n   - **Done** — bullets with commit hashes.\n   - **In flight** — file:line, what exists, what is missing.\n   - **Next steps** — ordered, each executable without re-deriving context\n     (exact commands, paths).\n   - **Verify** — the command(s) that prove the goal is met.\n   - **Context** — decisions already made; pointers to plan\u002Fspec\u002Ftask list.\n   - **Hops** — restart counter, `n`\u002F2. Incremented every time a restart is\n     scheduled; a stop phase that reads `2\u002F2` must not schedule again.\n4. **Report exact state.** The final message states what is committed\n   (hashes), what is NOT done, where the note is, and how to resume. Never\n   a formulation that reads as completion.\n5. **Offer one bounded restart.** The reset time is plan-specific and\n   unknowable from inside the session, so schedule conservatively and treat\n   the scheduled run itself as the probe: if it executes, the limit is gone;\n   if it cannot, nothing happens.\n   - **Host-native scheduler first.** ZCode: prefer the off-peak queue (no\n     plan-quota cost) when the work can wait; otherwise one one-shot\n     scheduled automation.\n   - **Otherwise, the OS scheduler driving the CLI's headless mode.** The\n     coding CLIs can continue a session non-interactively, so a local\n     one-shot cron\u002Flaunchd\u002Fsystemd entry works. Verified resume commands\n     (re-check `\u003Ccli> --help` before relying on a flag; they move between\n     versions):\n     - Claude Code: `claude --resume \u003Csession-id> -p \"\u003Cprompt>\"` (primary —\n       resumes exactly the interrupted session), falling back to\n       `claude -c -p \"\u003Cprompt>\"` (most recent session in this directory —\n       only safe when nothing else has run there since). Pass an unattended\n       permission set — `--permission-mode acceptEdits` plus a narrow\n       `--allowedTools` — never a blanket permission skip, and never a\n       blanket `Bash` allow: scope Bash to the command prefixes the resume\n       needs (`Bash(git:*)`, `Bash(python3 tests\u002F:*)`) plus the narrow\n       cleanup commands for this skill's own artifacts (`rm` of the wrapper\u002F\n       plist\u002Flog, `launchctl unload`, `crontab`).\n     - Codex: `codex exec resume --last \"\u003Cprompt>\"`.\n     - OpenCode: `opencode run -c \"\u003Cprompt>\"`.\n     - Cursor \u002F Augment \u002F Continue: no verified headless-resume path —\n       print exact resume instructions for the user instead of guessing.\n     Wrap the command in a small script kept **outside the repo** in a\n     per-user directory (e.g.\n     `\"${TMPDIR:-$HOME\u002F.cache}\u002Fsuperskills-quota\u002Fresume-\u003Crepo>.sh\"`, created\n     with `mkdir -p`; the `${TMPDIR:-\u002Ftmp}` fallback never lands in the\n     shared world-writable `\u002Ftmp`, so the tree stays clean and the file is\n     not swappable by another local account). The wrapper's **first action\n     is to remove its own scheduler entry** (unload + delete the plist,\n     strip the crontab line) — before it ever invokes the CLI — so the\n     one-shot guarantee never depends on the model run succeeding. It then\n     `cd`s into the repo and appends output to a log next to itself.\n     Register exactly **one** shot:\n     - Linux (and WSL): `systemd-run --user --on-active=\u003Cdelay>\n       --unit=quota-resume \u003Cscript>` — a transient timer, nothing to clean\n       up. Prefer this wherever it is available.\n     - macOS: a LaunchAgent plist under `~\u002FLibrary\u002FLaunchAgents` with a\n       `StartCalendarInterval` **pinning Month, Day, Hour and Minute** to a\n       concrete time `\u003Cdelay>` ahead — an interval naming only hour and\n       minute recurs *daily*, which is recurring spend, exactly what this\n       skill forbids. Do not rely on `at` — it is disabled by default on\n       macOS.\n     - Last resort: a crontab line tagged `# superskills-quota-resume`;\n       the resumed run strips exactly that line, preserving every other\n       entry — `crontab -l | grep -v '# superskills-quota-resume' | crontab -`\n       — never rewriting the crontab wholesale.\n   - **Consent:** a scheduled run is unattended model work — a spending\n     decision. Propose what will run, when, and roughly what it costs;\n     create it only after an explicit yes. The zero-cost off-peak path is\n     the exception.\n   - **Bounded:** one shot, never recurring. The scheduled prompt must\n     say: follow `\u002Fquota-resilience` — read `QUOTA-RESUME.md`, run the\n     Verify step, continue the Next steps in order, and if the limit hits\n     again, increment the note's **Hops** counter and schedule at most one\n     more hop; at `2\u002F2` it waits for the user instead.\n\n## Phase 2 — resuming after the stop\n\n1. Read `QUOTA-RESUME.md` first, before exploring the repo or re-deriving\n   state. It is the authority on where the last session stopped.\n2. Re-verify the cheapest way: run the note's Verify command before\n   building on the claimed state.\n3. Work the Next steps in order. On goal met + verification passing:\n   **delete the note** (its existence means \"interrupted\"), commit the\n   deletion, and remove any resume-scheduler artifacts you or the previous\n   session created (LaunchAgent plist, systemd unit, crontab line, wrapper\n   script and log).\n4. A new limit hit mid-resume → Phase 1 again, updating the existing note\n   rather than adding a second.\n\n## Boundaries\n\n- This skill is about the **agent's own** compute budget (usage limit,\n  quota exhaustion, hard stop). HTTP 429 \u002F rate limiting **received by the\n  user's application** is application behavior: `\u002Fdebug` or `\u002Fqa`, not this\n  skill.\n- Beyond the salvage commits above it mutates nothing, and it never pushes.\n","# \u002Fquota-resilience\n\nA usage-limit hard stop is a shutdown, not an obstacle to route around.\nRetrying the blocked call, downgrading the approach, or improvising \"one\nmore push\" spends the remaining window and makes the resume expensive. This\nskill spends that window making the resume cheap instead: committed work,\nclean tree, a note that carries the state, and one bounded restart where the\nhost can schedule one.\n\n## Phase 1 — on the limit error (stop)\n\nHard rules, in order:\n\n1. **Stop initiating work.** No retries of the blocked call, no new\n   subagents, no reduced-effort substitutes for the blocked task. Every\n   further call either fails identically or steals the window the shutdown\n   needs.\n2. **Salvage.** Split the working tree:\n   - *Complete* — passes its own check (tests, lint, build) or stands alone\n     (docs, config): commit with an honest message.\n   - *Incomplete* — never left dangling. Commit as a clearly labeled `wip:`\n     checkpoint if it is a useful base, or revert it if it is noise. Either\n     way `git status` ends **clean**: the next session starts from committed\n     state, never from a diff it has to reverse-engineer.\n   - Do not start new work to \"round out\" a commit.\n3. **Write the resume note** — `QUOTA-RESUME.md` at the repo root, committed\n   with the salvage. If this session's `handoff-\u003Csession-id>.md` (or a `handoff-*.md` the\n   work clearly belongs to) exists (written by `\u002Fhandoff` before the\n   stop), fold its content into this note and delete that file — one bridge,\n   not two. Leave other sessions' handoff files alone. Required sections:\n   - **Goal** — one line: the task's definition of done.\n   - **Done** — bullets with commit hashes.\n   - **In flight** — file:line, what exists, what is missing.\n   - **Next steps** — ordered, each executable without re-deriving context\n     (exact commands, paths).\n   - **Verify** — the command(s) that prove the goal is met.\n   - **Context** — decisions already made; pointers to plan\u002Fspec\u002Ftask list.\n   - **Hops** — restart counter, `n`\u002F2. Incremented every time a restart is\n     scheduled; a stop phase that reads `2\u002F2` must not schedule again.\n4. **Report exact state.** The final message states what is committed\n   (hashes), what is NOT done, where the note is, and how to resume. Never\n   a formulation that reads as completion.\n5. **Offer one bounded restart.** The reset time is plan-specific and\n   unknowable from inside the session, so schedule conservatively and treat\n   the scheduled run itself as the probe: if it executes, the limit is gone;\n   if it cannot, nothing happens.\n   - **Host-native scheduler first.** ZCode: prefer the off-peak queue (no\n     plan-quota cost) when the work can wait; otherwise one one-shot\n     scheduled automation.\n   - **Otherwise, the OS scheduler driving the CLI's headless mode.** The\n     coding CLIs can continue a session non-interactively, so a local\n     one-shot cron\u002Flaunchd\u002Fsystemd entry works. Verified resume commands\n     (re-check `\u003Ccli> --help` before relying on a flag; they move between\n     versions):\n     - Claude Code: `claude --resume \u003Csession-id> -p \"\u003Cprompt>\"` (primary —\n       resumes exactly the interrupted session), falling back to\n       `claude -c -p \"\u003Cprompt>\"` (most recent session in this directory —\n       only safe when nothing else has run there since). Pass an unattended\n       permission set — `--permission-mode acceptEdits` plus a narrow\n       `--allowedTools` — never a blanket permission skip, and never a\n       blanket `Bash` allow: scope Bash to the command prefixes the resume\n       needs (`Bash(git:*)`, `Bash(python3 tests\u002F:*)`) plus the narrow\n       cleanup commands for this skill's own artifacts (`rm` of the wrapper\u002F\n       plist\u002Flog, `launchctl unload`, `crontab`).\n     - Codex: `codex exec resume --last \"\u003Cprompt>\"`.\n     - OpenCode: `opencode run -c \"\u003Cprompt>\"`.\n     - Cursor \u002F Augment \u002F Continue: no verified headless-resume path —\n       print exact resume instructions for the user instead of guessing.\n     Wrap the command in a small script kept **outside the repo** in a\n     per-user directory (e.g.\n     `\"${TMPDIR:-$HOME\u002F.cache}\u002Fsuperskills-quota\u002Fresume-\u003Crepo>.sh\"`, created\n     with `mkdir -p`; the `${TMPDIR:-\u002Ftmp}` fallback never lands in the\n     shared world-writable `\u002Ftmp`, so the tree stays clean and the file is\n     not swappable by another local account). The wrapper's **first action\n     is to remove its own scheduler entry** (unload + delete the plist,\n     strip the crontab line) — before it ever invokes the CLI — so the\n     one-shot guarantee never depends on the model run succeeding. It then\n     `cd`s into the repo and appends output to a log next to itself.\n     Register exactly **one** shot:\n     - Linux (and WSL): `systemd-run --user --on-active=\u003Cdelay>\n       --unit=quota-resume \u003Cscript>` — a transient timer, nothing to clean\n       up. Prefer this wherever it is available.\n     - macOS: a LaunchAgent plist under `~\u002FLibrary\u002FLaunchAgents` with a\n       `StartCalendarInterval` **pinning Month, Day, Hour and Minute** to a\n       concrete time `\u003Cdelay>` ahead — an interval naming only hour and\n       minute recurs *daily*, which is recurring spend, exactly what this\n       skill forbids. Do not rely on `at` — it is disabled by default on\n       macOS.\n     - Last resort: a crontab line tagged `# superskills-quota-resume`;\n       the resumed run strips exactly that line, preserving every other\n       entry — `crontab -l | grep -v '# superskills-quota-resume' | crontab -`\n       — never rewriting the crontab wholesale.\n   - **Consent:** a scheduled run is unattended model work — a spending\n     decision. Propose what will run, when, and roughly what it costs;\n     create it only after an explicit yes. The zero-cost off-peak path is\n     the exception.\n   - **Bounded:** one shot, never recurring. The scheduled prompt must\n     say: follow `\u002Fquota-resilience` — read `QUOTA-RESUME.md`, run the\n     Verify step, continue the Next steps in order, and if the limit hits\n     again, increment the note's **Hops** counter and schedule at most one\n     more hop; at `2\u002F2` it waits for the user instead.\n\n## Phase 2 — resuming after the stop\n\n1. Read `QUOTA-RESUME.md` first, before exploring the repo or re-deriving\n   state. It is the authority on where the last session stopped.\n2. Re-verify the cheapest way: run the note's Verify command before\n   building on the claimed state.\n3. Work the Next steps in order. On goal met + verification passing:\n   **delete the note** (its existence means \"interrupted\"), commit the\n   deletion, and remove any resume-scheduler artifacts you or the previous\n   session created (LaunchAgent plist, systemd unit, crontab line, wrapper\n   script and log).\n4. A new limit hit mid-resume → Phase 1 again, updating the existing note\n   rather than adding a second.\n\n## Boundaries\n\n- This skill is about the **agent's own** compute budget (usage limit,\n  quota exhaustion, hard stop). HTTP 429 \u002F rate limiting **received by the\n  user's application** is application behavior: `\u002Fdebug` or `\u002Fqa`, not this\n  skill.\n- Beyond the salvage commits above it mutates nothing, and it never pushes.\n",{"name":5,"version":49,"triggers":50,"description":57,"allowed-tools":58},"1.1.0",[51,52,53,54,55,56],"hit my usage limit","usage limit hit","quota exhausted","quota hard stop","session died mid-task","resume after the limit resets","Salvage and resume when the session hits its own usage limit \u002F quota hard\nstop: commit complete work, leave a clean tree, write a committed resume\nnote, report exact state, and offer one bounded auto-restart where the host\ncan schedule one. Use when the agent's own usage limit, quota, or rate\nbudget is exhausted mid-task, or when picking up work a session left when\nit died that way. Not for HTTP 429s \u002F rate limits the user's application\nreceives — that is application behavior (\u002Fdebug, \u002Fqa), not agent state.\n",[28,59,60],"Read","Write","main","c7ec8d78e5dc37484c2cfe90ab04de163d97dd41","skills\u002Fquota-resilience",0,"https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills","https:\u002F\u002Fhyperion360.com\u002Fblog\u002Fparallel-ai-agents-engineering-workflow\u002F","https:\u002F\u002Fgithub.com\u002Fariadoss","https:\u002F\u002Fdanilosapad.com\u002F","CEO & Founder of @Hyperion360","2026-11-01 12:36:50",{"source":8,"skillName":5,"commands":72,"prompt":82,"zipAvailable":83},[73,76,79],{"tool":74,"command":75},"npx","npx skills add ariadoss\u002Fsuperskills --skill quota-resilience",{"tool":77,"command":78},"bunx","bunx skills add ariadoss\u002Fsuperskills --skill quota-resilience",{"tool":80,"command":81},"pnpm","pnpm dlx skills add ariadoss\u002Fsuperskills --skill quota-resilience","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills\u002Fblob\u002Fmain\u002Fskills\u002Fquota-resilience\u002FSKILL.md",true,[],[86,112,134,153,181,197],{"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":97,"publishedAt":44},"01M71AZE3MPF891RM354EMRYA1","youtube","Download YouTube video transcripts when user provides a YouTube URL or asks to download\u002Fget\u002Ffetch a transcript from YouTube. Also use when user wants to transcribe or get captions\u002Fsubtitles from a YouTube video.","https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube\u002FSKILL.md",[92],{"categoryNo":93,"categoryName":94,"taxonomyType":95,"taxonomyTypeDesc":96},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[88,98,99,100,101,102,103,28,104,29,105,106,107,108,109,32,110,111],"user","download","transcribe","write","transcript","video transcripts","URL","Captions","get","fetch","Video","read","subtitles","transcripts",{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"publishedAt":44},"01M71AZAYR85PW6BJGSQXS4BN2","write-plan","Use when you have a spec or requirements for a multi-step task, before touching code. Produces an engineering plan complete with TDD tasks, DRY\u002FSOLID\u002FYAGNI principles, a required Test Plan & Verification section, and a coverage target — then auto-runs \u002Fplan-eng-review as an independent double-check before execution.","https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills\u002Fblob\u002Fmain\u002Fskills\u002Fwrite-plan\u002FSKILL.md",[],[119,120,101,121,122,123,124,125,126,29,33,127,128,39,129,40,130,131,132,133],"write plan","plan","dry","verification","review","tasks","code","you","eng","coverage","section","test plan","target","as","yagni",{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":138,"categories":139,"keywords":140,"publishedAt":44},"01M71AZ8NDDP5ZMGEE6FSC4FKR","worktrees","Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification","https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills\u002Fblob\u002Fmain\u002Fskills\u002Fworktrees\u002FSKILL.md",[],[136,141,142,143,32,144,145,146,147,148,149,122,150,29,151,152],"implementation","workspace","with","plans","feature","isolation","git","directory","work","git worktrees","selection","Safety",{"skillNo":154,"name":155,"nameZh":6,"description":156,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":157,"categories":158,"keywords":162,"publishedAt":44},"01M71AZ4ZQHTNWFRZVYHKQFP08","web-perf","Analyzes web performance using Chrome DevTools MCP. Measures Core Web Vitals (LCP, INP, CLS) and supplementary metrics (FCP, TBT, Speed Index), identifies render-blocking resources, network dependency chains, layout shifts, caching issues, and accessibility gaps. Use when asked to audit, profile, debug, or optimize page load performance, Lighthouse scores, or site speed. Biases towards retrieval from current documentation over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills\u002Fblob\u002Fmain\u002Fskills\u002Fweb-perf\u002FSKILL.md",[159],{"categoryNo":160,"categoryName":161,"taxonomyType":95,"taxonomyTypeDesc":96},"01JBSEEDCAT000000000000023","Skill与智能体开发",[163,164,165,166,167,168,169,170,171,29,35,172,173,174,175,176,177,178,179,180],"web","perf","web perf","resources","MCP","layout","optimize","documentation","caching","knowledge","devtools","issues","render","re","performance","chrome devtools","vitals","never",{"skillNo":182,"name":183,"nameZh":6,"description":184,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":185,"categories":186,"keywords":187,"publishedAt":44},"01M71AZ2KZNWYEGKKRMJH8PYG2","verify","Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always","https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify\u002FSKILL.md",[],[183,39,188,189,149,122,190,191,29,192,32,193,194,195,196],"commands","about","claims","output","assertions","claim","prs","evidence","running",{"skillNo":198,"name":199,"nameZh":6,"description":200,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":201,"categories":202,"keywords":203,"publishedAt":44},"01M71AZ0ACCK7N1QMWQK1FV2HJ","validate-idea","Validate a business idea using the minimalist entrepreneur framework. Use when someone has a business idea and wants to test if it's worth pursuing before building anything.","https:\u002F\u002Fgithub.com\u002Fariadoss\u002Fsuperskills\u002Fblob\u002Fmain\u002Fskills\u002Fvalidate-idea\u002FSKILL.md",[],[204,205,206,207,208,209,210,29,211,32,212],"validate","idea","validate idea","framework","it","minimalist","test","business","before building"]