[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AGCXH3KE47TDMB7BA5TYWD":3,"skill-same:01M7AGCXH3KE47TDMB7BA5TYWD":84},{"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":6,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M7AGCXH3KE47TDMB7BA5TYWD","recallforge",null,"Turn raw exam materials—including PPT\u002FPPTX slides, digital or scanned PDFs, textbook scans, past papers, images, tables, formulas, diagrams, lecture notes, Markdown, and mixed Chinese-English course content—into source-grounded exam review with material inspection, knowledge reconstruction, active recall, weak-topic diagnosis, adaptive practice, study guides, and mock exams. Use for finals, midterms, revision, exam preparation, multimodal course-material review, or requests to inspect study files. Do not use for code or pull-request review, contract or product review, translation-only work, generic summarization without a learning goal, or unrelated development.","SiriZhao\u002Frecallforge-skill","SiriZhao","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F125785085?v=4","01M7AGCGEEXTB5A507WS1T3FDD","Siri Zhao","01M7AGCGF1NX35ECD7KHFY1HMV",0,"MIT","https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: recallforge\ndescription: Turn raw exam materials—including PPT\u002FPPTX slides, digital or scanned PDFs, textbook scans, past papers, images, tables, formulas, diagrams, lecture notes, Markdown, and mixed Chinese-English course content—into source-grounded exam review with material inspection, knowledge reconstruction, active recall, weak-topic diagnosis, adaptive practice, study guides, and mock exams. Use for finals, midterms, revision, exam preparation, multimodal course-material review, or requests to inspect study files. Do not use for code or pull-request review, contract or product review, translation-only work, generic summarization without a learning goal, or unrelated development.\n---\n\n# RecallForge — AI Exam Review Skill\n\nForge course materials into exam-ready knowledge. RecallForge is a host-executed learning workflow, not a standalone app, model, API service, or generic summarizer.\n\n## First response and mode selection\n\nIf the user says `self-test`, `run RecallForge self-test`, or `$recallforge self-test`, run the **Self-Test Mode** below exactly. Do not request uploads, API keys, or external tools.\n\nIf the user says `multimodal-test` or asks for the multimodal self-test, read [Material Intelligence](references\u002Fmaterial-intelligence.md) and run its multimodal self-test. If the host cannot inspect the bundled or attached asset, return `MULTIMODAL_HOST_CAPABILITY_UNAVAILABLE` rather than a false failure.\n\nIf the user says `inspect-materials` or asks to inspect files without starting review, read [Material Intelligence](references\u002Fmaterial-intelligence.md), build a truthful material catalog, report recognition quality, and stop before active recall.\n\nOtherwise, briefly identify the course, exam horizon if known, material supplied, and desired depth. Do not block on optional details. Use the material the user has actually provided; never claim to have opened or processed unavailable files.\n\nChoose the lightest appropriate mode:\n\n- **Quick review:** identify high-value ideas, ask a small active-recall round, then recommend one next step.\n- **Weakness mode:** test before explaining; review only demonstrated gaps.\n- **Deep review:** reconstruct a knowledge structure, work through it in small rounds, record stated weaknesses, and finish with a mock-exam option.\n- **Mock exam:** create questions only from supplied material; withhold answers until the user responds, then grade, diagnose, and target review.\n- **Cram mode:** use the stated time limit to prioritize retrieval and error repair over exhaustive explanation.\n\n## Core workflow\n\n1. **Material intelligence.** For supplied files, read [Material Intelligence](references\u002Fmaterial-intelligence.md). Catalog them, route each page\u002Fslide through reliable native extraction and host vision as needed, preserve source anchors and structure, and mark unresolved content. Never claim to inspect unavailable files.\n2. **Knowledge reconstruction.** Organize the material into a compact topic map: prerequisites, distinctions, high-value concepts, and evidence\u002Fsource anchors where available.\n3. **Exam prioritization.** Explain priorities using the material, past-paper evidence, teacher emphasis, exam date, and the learner's answers. Never invent exam frequency, teacher preferences, or readiness scores.\n4. **Active recall.** Ask one manageable question at a time by default. Let the learner answer before revealing the answer or explanation.\n5. **Weakness diagnosis.** Classify observed gaps (concept, condition, formula recall, method choice, calculation, interpretation, or careless error) and give the shortest repair action.\n6. **Targeted practice.** Generate a small follow-up question or drill from the supplied material. Keep it proportional to the learner's time.\n7. **Feedback loop.** State what changed, what remains uncertain, and the next review action. Treat mastery as unknown until the learner supplies real responses.\n\n## Output shape\n\nUse concise sections when useful:\n\n1. Knowledge structure\n2. Exam focus \u002F evidence boundary\n3. Active recall\n4. Practice or mock-exam item\n5. Weakness diagnosis (after an answer)\n6. Recommended next step\n\nDo not dump a long summary first. Prefer a map, a retrieval question, then an adaptive next step.\n\n## Self-Test Mode\n\nUse only [mini-course.md](assets\u002Fself-test\u002Fmini-course.md). Return one screen or less and include these exact status lines:\n\n```text\nRecallForge Self-Test\n✓ Skill activated\n✓ Course material parsed\n✓ Knowledge structure created\n✓ Active-recall question generated\n✓ Exam-style practice generated\nStatus: READY\n```\n\nThen list these detected topics: Conditional probability; Bayes' theorem; Independent events; Mutually exclusive events. Include one recall question asking for the difference between independent and mutually exclusive events, one short exam-style practice prompt, and one next-step recommendation. End with: `RecallForge is installed correctly. You can now attach your own course materials and start reviewing.`\n\n## Boundaries and integrity\n\n- Use only authorized course materials. Do not help obtain restricted exam content or facilitate cheating.\n- Do not fabricate citations, source coverage, past-paper frequency, teacher tendencies, readiness, or learner mastery.\n- Preserve formula conditions and terminology. Ask for a clearer source when a scan, table, diagram, handwriting, or formula is ambiguous.\n- Retain native text and visual evidence separately when they disagree. Never silently rewrite formulas, flatten tables, treat handwriting as an answer key, or drop an unresolved page.\n- Treat external knowledge as supplementary and label it as such when it is not supported by the supplied course material.\n- Support Chinese, English, and mixed-language material. Keep technical terms precise instead of blindly translating them.\n- Do not activate for ordinary code review, product review, legal review, translation-only work, or unrelated requests.\n\n## References\n\n- [Review methodology](references\u002Freview-methodology.md)\n- [Active recall](references\u002Factive-recall.md)\n- [Exam simulation](references\u002Fexam-simulation.md)\n- [Material Intelligence](references\u002Fmaterial-intelligence.md)\n","# RecallForge — AI Exam Review Skill\n\nForge course materials into exam-ready knowledge. RecallForge is a host-executed learning workflow, not a standalone app, model, API service, or generic summarizer.\n\n## First response and mode selection\n\nIf the user says `self-test`, `run RecallForge self-test`, or `$recallforge self-test`, run the **Self-Test Mode** below exactly. Do not request uploads, API keys, or external tools.\n\nIf the user says `multimodal-test` or asks for the multimodal self-test, read [Material Intelligence](references\u002Fmaterial-intelligence.md) and run its multimodal self-test. If the host cannot inspect the bundled or attached asset, return `MULTIMODAL_HOST_CAPABILITY_UNAVAILABLE` rather than a false failure.\n\nIf the user says `inspect-materials` or asks to inspect files without starting review, read [Material Intelligence](references\u002Fmaterial-intelligence.md), build a truthful material catalog, report recognition quality, and stop before active recall.\n\nOtherwise, briefly identify the course, exam horizon if known, material supplied, and desired depth. Do not block on optional details. Use the material the user has actually provided; never claim to have opened or processed unavailable files.\n\nChoose the lightest appropriate mode:\n\n- **Quick review:** identify high-value ideas, ask a small active-recall round, then recommend one next step.\n- **Weakness mode:** test before explaining; review only demonstrated gaps.\n- **Deep review:** reconstruct a knowledge structure, work through it in small rounds, record stated weaknesses, and finish with a mock-exam option.\n- **Mock exam:** create questions only from supplied material; withhold answers until the user responds, then grade, diagnose, and target review.\n- **Cram mode:** use the stated time limit to prioritize retrieval and error repair over exhaustive explanation.\n\n## Core workflow\n\n1. **Material intelligence.** For supplied files, read [Material Intelligence](references\u002Fmaterial-intelligence.md). Catalog them, route each page\u002Fslide through reliable native extraction and host vision as needed, preserve source anchors and structure, and mark unresolved content. Never claim to inspect unavailable files.\n2. **Knowledge reconstruction.** Organize the material into a compact topic map: prerequisites, distinctions, high-value concepts, and evidence\u002Fsource anchors where available.\n3. **Exam prioritization.** Explain priorities using the material, past-paper evidence, teacher emphasis, exam date, and the learner's answers. Never invent exam frequency, teacher preferences, or readiness scores.\n4. **Active recall.** Ask one manageable question at a time by default. Let the learner answer before revealing the answer or explanation.\n5. **Weakness diagnosis.** Classify observed gaps (concept, condition, formula recall, method choice, calculation, interpretation, or careless error) and give the shortest repair action.\n6. **Targeted practice.** Generate a small follow-up question or drill from the supplied material. Keep it proportional to the learner's time.\n7. **Feedback loop.** State what changed, what remains uncertain, and the next review action. Treat mastery as unknown until the learner supplies real responses.\n\n## Output shape\n\nUse concise sections when useful:\n\n1. Knowledge structure\n2. Exam focus \u002F evidence boundary\n3. Active recall\n4. Practice or mock-exam item\n5. Weakness diagnosis (after an answer)\n6. Recommended next step\n\nDo not dump a long summary first. Prefer a map, a retrieval question, then an adaptive next step.\n\n## Self-Test Mode\n\nUse only [mini-course.md](assets\u002Fself-test\u002Fmini-course.md). Return one screen or less and include these exact status lines:\n\n```text\nRecallForge Self-Test\n✓ Skill activated\n✓ Course material parsed\n✓ Knowledge structure created\n✓ Active-recall question generated\n✓ Exam-style practice generated\nStatus: READY\n```\n\nThen list these detected topics: Conditional probability; Bayes' theorem; Independent events; Mutually exclusive events. Include one recall question asking for the difference between independent and mutually exclusive events, one short exam-style practice prompt, and one next-step recommendation. End with: `RecallForge is installed correctly. You can now attach your own course materials and start reviewing.`\n\n## Boundaries and integrity\n\n- Use only authorized course materials. Do not help obtain restricted exam content or facilitate cheating.\n- Do not fabricate citations, source coverage, past-paper frequency, teacher tendencies, readiness, or learner mastery.\n- Preserve formula conditions and terminology. Ask for a clearer source when a scan, table, diagram, handwriting, or formula is ambiguous.\n- Retain native text and visual evidence separately when they disagree. Never silently rewrite formulas, flatten tables, treat handwriting as an answer key, or drop an unresolved page.\n- Treat external knowledge as supplementary and label it as such when it is not supported by the supplied course material.\n- Support Chinese, English, and mixed-language material. Keep technical terms precise instead of blindly translating them.\n- Do not activate for ordinary code review, product review, legal review, translation-only work, or unrelated requests.\n\n## References\n\n- [Review methodology](references\u002Freview-methodology.md)\n- [Active recall](references\u002Factive-recall.md)\n- [Exam simulation](references\u002Fexam-simulation.md)\n- [Material Intelligence](references\u002Fmaterial-intelligence.md)\n",{"name":5,"description":7},"main","0e648f61655642250c2562d66c3ef55de4194602","skill\u002Frecallforge","https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill","https:\u002F\u002Fgithub.com\u002FSiriZhao","A student who eager to learn from peers, committed to open-source and community contributions.","2026-08-13 21:34:03",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add SiriZhao\u002Frecallforge-skill --skill recallforge",{"tool":37,"command":38},"bunx","bunx skills add SiriZhao\u002Frecallforge-skill --skill recallforge",{"tool":40,"command":41},"pnpm","pnpm dlx skills add SiriZhao\u002Frecallforge-skill --skill recallforge","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002FSKILL.md",true,[45,51,57,60,64,68,72,76,80],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"agents\u002Fopenai.yaml","REFERENCE","参考文档",418,"https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Fagents\u002Fopenai.yaml",{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},"assets\u002Frecallforge-banner.svg","ASSET","资产文件",682,"https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Fassets\u002Frecallforge-banner.svg",{"path":58,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":49,"sourceHtmlUrl":59},"assets\u002Frecallforge-mark.svg","https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Fassets\u002Frecallforge-mark.svg",{"path":61,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":62,"sourceHtmlUrl":63},"assets\u002Fself-test\u002Fmini-course.md",321,"https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Fassets\u002Fself-test\u002Fmini-course.md",{"path":65,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":66,"sourceHtmlUrl":67},"assets\u002Fself-test\u002Fmultimodal\u002Fprobability-slide.svg",2613,"https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Fassets\u002Fself-test\u002Fmultimodal\u002Fprobability-slide.svg",{"path":69,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":70,"sourceHtmlUrl":71},"references\u002Factive-recall.md",485,"https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Freferences\u002Factive-recall.md",{"path":73,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Fexam-simulation.md",433,"https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Freferences\u002Fexam-simulation.md",{"path":77,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fmaterial-intelligence.md",4506,"https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Freferences\u002Fmaterial-intelligence.md",{"path":81,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Freview-methodology.md",667,"https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Fskill\u002Frecallforge\u002Freferences\u002Freview-methodology.md",[85,90],{"skillNo":86,"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":87,"categories":88,"keywords":89,"publishedAt":19},"01M7AGCQK9XZTH38XMSS53RVS0","https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002Frecallforge-plugin\u002Fskills\u002Frecallforge\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":93,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M7AGCKQ80M8P468MNBFQFP0J","recallforge-skill","https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FSiriZhao\u002Frecallforge-skill\u002Fblob\u002Fmain\u002FSKILL.md",[],[]]