[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7860WSCQMXFD45FTBCEEPG3":3,"skill-same:01M7860WSCQMXFD45FTBCEEPG3":43},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":13,"repositoryHtmlUrl":26,"repositoryHomepageUrl":6,"ownerHtmlUrl":27,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":28,"install":29,"resources":42},"01M7860WSCQMXFD45FTBCEEPG3","up-snapshot",null,"Export current project state as a portable markdown snapshot for external research","cvedim9\u002FHPRP-audit","cvedim9","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F177762869?v=4","01M785Z4QRNA6FNQBQ3M43BPSN","01M785Z4RPAZZ36PHSHGS7RBWB",0,"MIT","https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit\u002Fblob\u002Fmain\u002F.unpack\u002Fskills\u002Fup-snapshot\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: up-snapshot\ndescription: Export current project state as a portable markdown snapshot for external research\n---\n\n# Export project state as a portable snapshot\n\nCreate a narrative markdown snapshot of the project that can be pasted into ChatGPT, Claude, or any AI tool for research conversations.\n\n## Input\n\n`$ARGUMENTS` may contain a description of what the user wants (e.g., \"database schemas\", \"full overview\"). If empty, ask.\n\n## Step 1: Figure out what they need\n\nIf `$ARGUMENTS` already describes a clear intent, skip to Step 2.\n\nOtherwise, present this message (keep it exactly this brief):\n\n```\nWhat kind of snapshot?\n\n- **Overview** — what this project is, where we are, what's decided\n- **Design focus** — specific area you want to discuss (e.g., \"redesign the UI\", \"rethink the data model\")\n- **Implementation** — how something actually works today, with schemas and code references\n- **Everything** — full dump, no filter\n\nOr just tell me what you want to take to ChatGPT and I'll figure it out.\n```\n\nWait for the user's response.\n\n**If they pick design focus or implementation but don't say which area:** ask \"Which area?\" — nothing else. Wait again. That's your max 2 rounds of clarification.\n\n**If they describe something specific in any form:** go straight to Step 2. Don't ask for clarification you don't need.\n\n## Step 2: Read project state\n\nRead what you need based on the intent — not everything:\n\n- **Overview**: `.unpack\u002Fdocs\u002Findex.md`, `.unpack\u002Fdocs\u002Fspecs\u002F*`, `.unpack\u002Fdocs\u002F_meta\u002Fproject-memory.md`\n- **Design focus**: `.unpack\u002Fdocs\u002Findex.md`, the relevant specs, the relevant phases, relevant decisions from `project-memory.md`\n- **Implementation**: `.unpack\u002Fdocs\u002Findex.md`, relevant specs, relevant phases, AND relevant source files (schemas, configs, key modules) — yes, include actual code for implementation snapshots\n- **Everything**: all docs (`index.md`, `specs\u002F*`, `phases\u002F*`, `_meta\u002Fproject-memory.md`, `practices\u002F*` under `.unpack\u002Fdocs\u002F`), conversation list from `.unpack\u002Fconversations\u002F`\n\nUse Task agents with subagent_type=Explore if you need to find relevant source files for implementation snapshots. For other types, direct reads are fine.\n\n## Step 3: Write the snapshot\n\nWrite to `snapshot.md` at the repo root.\n\n**The snapshot must be narrative, not a file listing.** Write about the project as if you're briefing someone who will help you think about it. Reference file paths inline only when they ground the narrative (e.g., \"the claim extraction pipeline (`packages\u002Fdb\u002Fscripts\u002Fextract-claims.ts`) reads from...\").\n\n### Structure by snapshot type\n\n**Overview:**\n```markdown\n# \u003CProject Name> — Snapshot\n\n> \u003C1-sentence description of the project>\n\n## Where we are\n\u003CPhase status, what's done, what's in progress, what's next. Narrative, not a table.>\n\n## How it works\n\u003CArchitecture, components, data flows — described narratively with domain language.>\n\n## Domain model\n\u003CKey entities, relationships, business rules. Use the project's own terminology.>\n\n## Key decisions\n\u003COnly the decisions that shape the project's identity. Include rationale.>\n\n## Open questions\n\u003CWhat's unresolved, what needs input.>\n```\n\n**Design focus:**\n```markdown\n# \u003CProject Name> — \u003CArea> Design Snapshot\n\n> \u003C1-sentence project context + what this snapshot focuses on>\n\n## Project context\n\u003CJust enough to understand the project — 1-2 paragraphs max.>\n\n## Current state of \u003Carea>\n\u003CHow it works today. What's built, what's not. Include relevant schemas, patterns, or code snippets.>\n\n## Decisions shaping \u003Carea>\n\u003COnly the decisions relevant to this area, with rationale.>\n\n## What we want to discuss\n\u003CRestate the user's intent — what they want to explore, redesign, or decide.>\n\n## Open questions\n\u003CUnresolved items relevant to this area.>\n```\n\n**Implementation:**\n```markdown\n# \u003CProject Name> — \u003CArea> Implementation Snapshot\n\n> \u003C1-sentence context>\n\n## How \u003Carea> works\n\u003CStep-by-step narrative of the current implementation. Include actual code snippets, schema definitions, config shapes — whatever makes it concrete.>\n\n## Key files\n\u003CBrief map of where things live, only for the relevant area.>\n\n## Decisions and constraints\n\u003CTechnical decisions that constrain this area.>\n\n## Known issues \u002F gaps\n\u003CWhat's missing, broken, or needs improvement.>\n```\n\n**Everything:**\nUse the Overview structure but expand every section fully. Include phase details, all decisions, coding standards summary, conversation history.\n\n### Writing rules\n\n- **Speak about the project, not about files.** \"The system extracts claims from transcribed speech\" not \"The extraction spec is in `.unpack\u002Fdocs\u002Fspecs\u002F05-extraction.md`\"\n- **Use the project's domain language.** If the project calls things \"episodes\" and \"claims\", use those words — the point is that someone reading this snapshot will speak the same language when they respond\n- **Include code\u002Fschemas when they add clarity.** Especially for implementation snapshots — paste the actual Prisma schema, the actual API route, the actual config shape\n- **Inline file references where useful.** \"(see `packages\u002Fdb\u002Fprisma\u002Fschema.prisma`)\" — but only when someone might want to find it later\n- **Keep under 8000 words** if possible — this goes into a context window\n\n## Step 4: Report\n\nAfter writing the snapshot, say:\n\n```\nSnapshot saved to `snapshot.md`.\n\nPaste it into ChatGPT or Claude to continue your research. When you're done, save the conversation as `conversation.md` and run `\u002Fup-apply`.\n```\n\nThat's it. Two lines.\n\n## Important\n\n- Keep your messages to the user SHORT — 1-2 lines unless presenting a list\n- Do NOT read files before asking the user what they want\n- Do NOT use AskUserQuestion — this is conversational\n- Max 3 rounds of back-and-forth before you start building\n- If the user gives you enough info in `$ARGUMENTS`, skip straight to reading + writing (0 rounds)\n","# Export project state as a portable snapshot\n\nCreate a narrative markdown snapshot of the project that can be pasted into ChatGPT, Claude, or any AI tool for research conversations.\n\n## Input\n\n`$ARGUMENTS` may contain a description of what the user wants (e.g., \"database schemas\", \"full overview\"). If empty, ask.\n\n## Step 1: Figure out what they need\n\nIf `$ARGUMENTS` already describes a clear intent, skip to Step 2.\n\nOtherwise, present this message (keep it exactly this brief):\n\n```\nWhat kind of snapshot?\n\n- **Overview** — what this project is, where we are, what's decided\n- **Design focus** — specific area you want to discuss (e.g., \"redesign the UI\", \"rethink the data model\")\n- **Implementation** — how something actually works today, with schemas and code references\n- **Everything** — full dump, no filter\n\nOr just tell me what you want to take to ChatGPT and I'll figure it out.\n```\n\nWait for the user's response.\n\n**If they pick design focus or implementation but don't say which area:** ask \"Which area?\" — nothing else. Wait again. That's your max 2 rounds of clarification.\n\n**If they describe something specific in any form:** go straight to Step 2. Don't ask for clarification you don't need.\n\n## Step 2: Read project state\n\nRead what you need based on the intent — not everything:\n\n- **Overview**: `.unpack\u002Fdocs\u002Findex.md`, `.unpack\u002Fdocs\u002Fspecs\u002F*`, `.unpack\u002Fdocs\u002F_meta\u002Fproject-memory.md`\n- **Design focus**: `.unpack\u002Fdocs\u002Findex.md`, the relevant specs, the relevant phases, relevant decisions from `project-memory.md`\n- **Implementation**: `.unpack\u002Fdocs\u002Findex.md`, relevant specs, relevant phases, AND relevant source files (schemas, configs, key modules) — yes, include actual code for implementation snapshots\n- **Everything**: all docs (`index.md`, `specs\u002F*`, `phases\u002F*`, `_meta\u002Fproject-memory.md`, `practices\u002F*` under `.unpack\u002Fdocs\u002F`), conversation list from `.unpack\u002Fconversations\u002F`\n\nUse Task agents with subagent_type=Explore if you need to find relevant source files for implementation snapshots. For other types, direct reads are fine.\n\n## Step 3: Write the snapshot\n\nWrite to `snapshot.md` at the repo root.\n\n**The snapshot must be narrative, not a file listing.** Write about the project as if you're briefing someone who will help you think about it. Reference file paths inline only when they ground the narrative (e.g., \"the claim extraction pipeline (`packages\u002Fdb\u002Fscripts\u002Fextract-claims.ts`) reads from...\").\n\n### Structure by snapshot type\n\n**Overview:**\n```markdown\n# \u003CProject Name> — Snapshot\n\n> \u003C1-sentence description of the project>\n\n## Where we are\n\u003CPhase status, what's done, what's in progress, what's next. Narrative, not a table.>\n\n## How it works\n\u003CArchitecture, components, data flows — described narratively with domain language.>\n\n## Domain model\n\u003CKey entities, relationships, business rules. Use the project's own terminology.>\n\n## Key decisions\n\u003COnly the decisions that shape the project's identity. Include rationale.>\n\n## Open questions\n\u003CWhat's unresolved, what needs input.>\n```\n\n**Design focus:**\n```markdown\n# \u003CProject Name> — \u003CArea> Design Snapshot\n\n> \u003C1-sentence project context + what this snapshot focuses on>\n\n## Project context\n\u003CJust enough to understand the project — 1-2 paragraphs max.>\n\n## Current state of \u003Carea>\n\u003CHow it works today. What's built, what's not. Include relevant schemas, patterns, or code snippets.>\n\n## Decisions shaping \u003Carea>\n\u003COnly the decisions relevant to this area, with rationale.>\n\n## What we want to discuss\n\u003CRestate the user's intent — what they want to explore, redesign, or decide.>\n\n## Open questions\n\u003CUnresolved items relevant to this area.>\n```\n\n**Implementation:**\n```markdown\n# \u003CProject Name> — \u003CArea> Implementation Snapshot\n\n> \u003C1-sentence context>\n\n## How \u003Carea> works\n\u003CStep-by-step narrative of the current implementation. Include actual code snippets, schema definitions, config shapes — whatever makes it concrete.>\n\n## Key files\n\u003CBrief map of where things live, only for the relevant area.>\n\n## Decisions and constraints\n\u003CTechnical decisions that constrain this area.>\n\n## Known issues \u002F gaps\n\u003CWhat's missing, broken, or needs improvement.>\n```\n\n**Everything:**\nUse the Overview structure but expand every section fully. Include phase details, all decisions, coding standards summary, conversation history.\n\n### Writing rules\n\n- **Speak about the project, not about files.** \"The system extracts claims from transcribed speech\" not \"The extraction spec is in `.unpack\u002Fdocs\u002Fspecs\u002F05-extraction.md`\"\n- **Use the project's domain language.** If the project calls things \"episodes\" and \"claims\", use those words — the point is that someone reading this snapshot will speak the same language when they respond\n- **Include code\u002Fschemas when they add clarity.** Especially for implementation snapshots — paste the actual Prisma schema, the actual API route, the actual config shape\n- **Inline file references where useful.** \"(see `packages\u002Fdb\u002Fprisma\u002Fschema.prisma`)\" — but only when someone might want to find it later\n- **Keep under 8000 words** if possible — this goes into a context window\n\n## Step 4: Report\n\nAfter writing the snapshot, say:\n\n```\nSnapshot saved to `snapshot.md`.\n\nPaste it into ChatGPT or Claude to continue your research. When you're done, save the conversation as `conversation.md` and run `\u002Fup-apply`.\n```\n\nThat's it. Two lines.\n\n## Important\n\n- Keep your messages to the user SHORT — 1-2 lines unless presenting a list\n- Do NOT read files before asking the user what they want\n- Do NOT use AskUserQuestion — this is conversational\n- Max 3 rounds of back-and-forth before you start building\n- If the user gives you enough info in `$ARGUMENTS`, skip straight to reading + writing (0 rounds)\n",{"name":5,"description":7},"main","cde3f1fef7b187d731089c9d00ab656830139521",".unpack\u002Fskills\u002Fup-snapshot","https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit","https:\u002F\u002Fgithub.com\u002Fcvedim9","2026-07-06 10:44:49",{"source":8,"skillName":5,"commands":30,"prompt":40,"zipAvailable":41},[31,34,37],{"tool":32,"command":33},"npx","npx skills add cvedim9\u002FHPRP-audit --skill up-snapshot",{"tool":35,"command":36},"bunx","bunx skills add cvedim9\u002FHPRP-audit --skill up-snapshot",{"tool":38,"command":39},"pnpm","pnpm dlx skills add cvedim9\u002FHPRP-audit --skill up-snapshot","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit\u002Fblob\u002Fmain\u002F.unpack\u002Fskills\u002Fup-snapshot\u002FSKILL.md",true,[],[44,51,58,65,72,79],{"skillNo":45,"name":46,"nameZh":6,"description":47,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":48,"categories":49,"keywords":50,"publishedAt":18},"01M7860YR37HE6DEHABDX721DH","up-status","Show project progress, phases, blockers, and the next runnable phase","https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit\u002Fblob\u002Fmain\u002F.unpack\u002Fskills\u002Fup-status\u002FSKILL.md",[],[],{"skillNo":52,"name":53,"nameZh":6,"description":54,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":18},"01M7860TV6RN4TPJVP2NB6ZSC8","up-next","Execute the next runnable build phase","https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit\u002Fblob\u002Fmain\u002F.unpack\u002Fskills\u002Fup-next\u002FSKILL.md",[],[],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":18},"01M7860S519NNE58JZCRS8TGCR","up-init","Initialize a new project from the Unpack template","https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit\u002Fblob\u002Fmain\u002F.unpack\u002Fskills\u002Fup-init\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M7860PJKZF3Y3975XDE3QGRJ","up-extract-standards","Scan an example repo and extract its coding patterns into a standards file","https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit\u002Fblob\u002Fmain\u002F.unpack\u002Fskills\u002Fup-extract-standards\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M7860KP8NENS8G3BV9MZX41G","up-document","Generate or update human-readable docs in guide\u002F (supports Mintlify)","https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit\u002Fblob\u002Fmain\u002F.unpack\u002Fskills\u002Fup-document\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M7860HJ0TVHY2NS75Q4E8F35","up-bootstrap","Process a research conversation into structured docs, specs, and phases","https:\u002F\u002Fgithub.com\u002Fcvedim9\u002FHPRP-audit\u002Fblob\u002Fmain\u002F.unpack\u002Fskills\u002Fup-bootstrap\u002FSKILL.md",[],[]]