[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M77XGBC08DP018FQY16FM0X8":3,"skill-same:01M77XGBC08DP018FQY16FM0X8":67},{"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":29,"commitSha":30,"skillDirectory":31,"forksCount":32,"repositoryHtmlUrl":33,"repositoryHomepageUrl":34,"ownerHtmlUrl":35,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":36,"install":37,"resources":50},"01M77XGBC08DP018FQY16FM0X8","papers",null,"Search preprints from arXiv, bioRxiv, and medRxiv without API keys. Supports search, lookup, extraction, citations, enrichment, and paper downloads that bundle PDFs plus linked GitHub code.","StartupBros-com\u002Fhov-stack","StartupBros-com","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F262958016?v=4","01M77WY8AH50FY939DTGZYHVE9","StartupBros.com","01M77WY8B5Q2D7VSS0P6Y0JX6E",1,"MIT","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-patches\u002Fpapers\u002F.base\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: papers\ndescription: Search preprints from arXiv, bioRxiv, and medRxiv without API keys. Supports search, lookup, extraction, citations, enrichment, and paper downloads that bundle PDFs plus linked GitHub code.\ncompatibility: Optimized for Claude Code. Install in project .claude\u002Fskills\u002F or personal\n  ~\u002F.claude\u002Fskills\u002F. Pure local Python with no servers, no MCP, and no API keys.\nallowed-tools: Read Write Edit Grep Glob Bash Python\nmetadata:\n  short-description: Download preprints as paper\u002Fcode bundles with auto-cloned GitHub repos\nlicense: MIT\ndistribution: public\n---\n\n# Papers\n\n## Purpose\n\nSearch, retrieve, download, extract, analyze, and synthesize preprint papers\nfrom arXiv, bioRxiv, and medRxiv using pure local Python with no API keys.\nPrefer bundled scripts over ad hoc scraping: they return stable JSON,\ndeduplicate cross-source results, and handle source-specific URL quirks.\n\n## Workflow\n\n1. Run `search` first unless the user already has an arXiv ID or DOI.\n2. Read the JSON results and choose the highest-signal papers. Keep\n   `source`, `source_id`, `pdf_url`, and `text_url`.\n3. Read `retrieval_capabilities` before fetching. Use `blocked_reason` and\n   `section_extraction_quality` instead of assuming every URL works.\n4. Prefer `text_url` over `pdf_url` when it exists. XML extraction is cleaner.\n5. Use `download` only when a local copy is needed in the workspace.\n   Paper-aware downloads may also clone a linked public code repository into\n   the bundle's `code\u002F` directory when one is detected. Before running\n   `download` on a record with a `repo_url`, tell the user that code will be\n   cloned from the public host and confirm they want the code copy. Treat cloned\n   code as untrusted source material: read it, do not execute it unless the user\n   separately asks for that and the repo has been reviewed.\n6. Use `extract` to produce full text or specific sections for review.\n7. Use `analyze` to run an ambiguity audit on methodological gaps.\n8. Use `synthesize` for creative multi-query search and literature synthesis.\n9. Use `workflow` for the full search-to-synthesis pipeline in one call.\n10. Explicitly note that preprints are not peer reviewed.\n\n## Script Reference\n\n### Core CLI (`paper_search.py`)\n\n- `search` — Multi-source topic search. arXiv supports direct queries;\n  bioRxiv\u002FmedRxiv search recent windows and rank locally.\n- `lookup` — Resolve a known arXiv ID or DOI with Crossref fallback.\n- `download` — Fetch any `pdf_url` or `text_url` to a task-local directory.\n  When downloading by paper identifier or saved record, it creates a paper bundle\n  directory with `paper\u002F` and `code\u002F` subdirectories, stores the paper asset in\n  `paper\u002F`, and clones the detected repository into `code\u002F` when a GitHub-style\n  repo URL is available. Confirm before cloning and never execute cloned code by\n  default.\n- `extract` — Extract full text or named sections from XML, TXT, or PDF.\n  Section extraction is strongest on JATS XML, medium on plain text,\n  best-effort on PDF.\n- `analyze` — Run the ambiguity audit against either a raw sections map or an\n  `extract --json` payload plus paper metadata.\n- `synthesize` — Search and synthesize preprint literature. Use `--creative`\n  when you want query expansion instead of standard search.\n- `workflow` — Full pipeline: search → optional download → extract → optional\n  audit → synthesis → citations. `--deep-analyze` works even without\n  `--download`.\n- `cite` — Generate BibTeX, APA, or RIS citations from a record or identifier.\n- `enrich-related` — Fetch similar arXiv papers plus citation-graph links.\n- `extract-tikz` — Extract TikZ figures from arXiv source archives.\n- `doctor` — Check curl, pdftotext, cache writability, and provider reachability.\n\n### Standalone Utilities\n\n- `analysis.py`, `synthesis.py`, and `workflow.py` remain available as direct\n  scripts, but `paper_search.py` is the authoritative CLI surface.\n\n## Command Rules\n\n- `search`: Use for topic, author, and category discovery. Add `--days` for\n  tighter windows on broad biomedical queries. Use `--sort date` for recency.\n- `lookup`: Use when the user has an arXiv ID or DOI. Falls back to Crossref.\n- `download`: Keep downloads in a task-local directory, not the repo root.\n  Prefer `download --source ... --id ...` or `download --record ...` for\n  paper-aware bundles; use `download --url ...` only for raw asset fetches.\n  If the selected record has `repo_url`, the paper-aware bundle may run\n  `git clone --depth 1` for that public repo. Ask before the clone and use\n  `--repo-discovery off` during lookup\u002Fsearch when the user wants paper-only\n  retrieval.\n- `extract`: Use `--write` for long output. Use `--sections` for named sections.\n  Use `--json` for structured inline results.\n- `analyze`: Pass extracted sections JSON or `extract --json` output plus paper\n  metadata. Returns flags\n  with citation anchors (section + paragraph index).\n- `synthesize`: Pass a query for standard or creative search, or pass paper\n  records for literature synthesis. Groups papers by theme.\n- `workflow`: Pass `--creative` for query expansion, `--download` for PDFs,\n  `--deep-analyze` for ambiguity audit, and `--output-dir` if you want\n  downloaded assets persisted. Combines all steps.\n\n## Output Contract\n\nEvery search result includes `retrieval_capabilities` with\n`can_fetch_pdf`, `can_fetch_text`, `text_format`,\n`section_extraction_quality`, and `blocked_reason`.\n\nErrors emit JSON on stderr with `error_code`, `message`, `retryable`,\n`source`, `http_status`, and `details`.\n\nAudit flags include `category`, `status`, `anchor`, and `description`.\nWorkflow results may include `warnings` for degraded-but-nonfatal provider,\ndownload, extraction, or citation paths.\n\n## Source-Specific Guidance\n\n- arXiv is strongest for CS, ML, math, physics, quantitative work.\n- bioRxiv and medRxiv are strongest for life sciences and medicine.\n- When a query spans ML plus biology\u002Fmedicine, search all three.\n- For broad biomedical queries, reduce `--days` before increasing `--limit`.\n- medRxiv content downloads may return 403 in some environments.\n- arXiv enforces polite access; the script backs off automatically.\n\n## Guardrails\n\n- Never claim a preprint is peer-reviewed.\n- Respect arXiv 3-second polite delay (enforced in paper_search_lib).\n- Do not create files inside the skill directory; use task-local paths.\n- Ask for clarification when the topic is missing or ambiguous.\n- When a source is unavailable, degrade gracefully and report what succeeded.\n\n## Adapter Notes\n\n- Prefer bundled scripts over ad hoc shell commands for search, download, and extraction.\n- Keep the authored source of truth in split project files, not in generated adapter bundles.\n- All results are preprints. Never claim peer-review status.\n","# Papers\n\n## Purpose\n\nSearch, retrieve, download, extract, analyze, and synthesize preprint papers\nfrom arXiv, bioRxiv, and medRxiv using pure local Python with no API keys.\nPrefer bundled scripts over ad hoc scraping: they return stable JSON,\ndeduplicate cross-source results, and handle source-specific URL quirks.\n\n## Workflow\n\n1. Run `search` first unless the user already has an arXiv ID or DOI.\n2. Read the JSON results and choose the highest-signal papers. Keep\n   `source`, `source_id`, `pdf_url`, and `text_url`.\n3. Read `retrieval_capabilities` before fetching. Use `blocked_reason` and\n   `section_extraction_quality` instead of assuming every URL works.\n4. Prefer `text_url` over `pdf_url` when it exists. XML extraction is cleaner.\n5. Use `download` only when a local copy is needed in the workspace.\n   Paper-aware downloads may also clone a linked public code repository into\n   the bundle's `code\u002F` directory when one is detected. Before running\n   `download` on a record with a `repo_url`, tell the user that code will be\n   cloned from the public host and confirm they want the code copy. Treat cloned\n   code as untrusted source material: read it, do not execute it unless the user\n   separately asks for that and the repo has been reviewed.\n6. Use `extract` to produce full text or specific sections for review.\n7. Use `analyze` to run an ambiguity audit on methodological gaps.\n8. Use `synthesize` for creative multi-query search and literature synthesis.\n9. Use `workflow` for the full search-to-synthesis pipeline in one call.\n10. Explicitly note that preprints are not peer reviewed.\n\n## Script Reference\n\n### Core CLI (`paper_search.py`)\n\n- `search` — Multi-source topic search. arXiv supports direct queries;\n  bioRxiv\u002FmedRxiv search recent windows and rank locally.\n- `lookup` — Resolve a known arXiv ID or DOI with Crossref fallback.\n- `download` — Fetch any `pdf_url` or `text_url` to a task-local directory.\n  When downloading by paper identifier or saved record, it creates a paper bundle\n  directory with `paper\u002F` and `code\u002F` subdirectories, stores the paper asset in\n  `paper\u002F`, and clones the detected repository into `code\u002F` when a GitHub-style\n  repo URL is available. Confirm before cloning and never execute cloned code by\n  default.\n- `extract` — Extract full text or named sections from XML, TXT, or PDF.\n  Section extraction is strongest on JATS XML, medium on plain text,\n  best-effort on PDF.\n- `analyze` — Run the ambiguity audit against either a raw sections map or an\n  `extract --json` payload plus paper metadata.\n- `synthesize` — Search and synthesize preprint literature. Use `--creative`\n  when you want query expansion instead of standard search.\n- `workflow` — Full pipeline: search → optional download → extract → optional\n  audit → synthesis → citations. `--deep-analyze` works even without\n  `--download`.\n- `cite` — Generate BibTeX, APA, or RIS citations from a record or identifier.\n- `enrich-related` — Fetch similar arXiv papers plus citation-graph links.\n- `extract-tikz` — Extract TikZ figures from arXiv source archives.\n- `doctor` — Check curl, pdftotext, cache writability, and provider reachability.\n\n### Standalone Utilities\n\n- `analysis.py`, `synthesis.py`, and `workflow.py` remain available as direct\n  scripts, but `paper_search.py` is the authoritative CLI surface.\n\n## Command Rules\n\n- `search`: Use for topic, author, and category discovery. Add `--days` for\n  tighter windows on broad biomedical queries. Use `--sort date` for recency.\n- `lookup`: Use when the user has an arXiv ID or DOI. Falls back to Crossref.\n- `download`: Keep downloads in a task-local directory, not the repo root.\n  Prefer `download --source ... --id ...` or `download --record ...` for\n  paper-aware bundles; use `download --url ...` only for raw asset fetches.\n  If the selected record has `repo_url`, the paper-aware bundle may run\n  `git clone --depth 1` for that public repo. Ask before the clone and use\n  `--repo-discovery off` during lookup\u002Fsearch when the user wants paper-only\n  retrieval.\n- `extract`: Use `--write` for long output. Use `--sections` for named sections.\n  Use `--json` for structured inline results.\n- `analyze`: Pass extracted sections JSON or `extract --json` output plus paper\n  metadata. Returns flags\n  with citation anchors (section + paragraph index).\n- `synthesize`: Pass a query for standard or creative search, or pass paper\n  records for literature synthesis. Groups papers by theme.\n- `workflow`: Pass `--creative` for query expansion, `--download` for PDFs,\n  `--deep-analyze` for ambiguity audit, and `--output-dir` if you want\n  downloaded assets persisted. Combines all steps.\n\n## Output Contract\n\nEvery search result includes `retrieval_capabilities` with\n`can_fetch_pdf`, `can_fetch_text`, `text_format`,\n`section_extraction_quality`, and `blocked_reason`.\n\nErrors emit JSON on stderr with `error_code`, `message`, `retryable`,\n`source`, `http_status`, and `details`.\n\nAudit flags include `category`, `status`, `anchor`, and `description`.\nWorkflow results may include `warnings` for degraded-but-nonfatal provider,\ndownload, extraction, or citation paths.\n\n## Source-Specific Guidance\n\n- arXiv is strongest for CS, ML, math, physics, quantitative work.\n- bioRxiv and medRxiv are strongest for life sciences and medicine.\n- When a query spans ML plus biology\u002Fmedicine, search all three.\n- For broad biomedical queries, reduce `--days` before increasing `--limit`.\n- medRxiv content downloads may return 403 in some environments.\n- arXiv enforces polite access; the script backs off automatically.\n\n## Guardrails\n\n- Never claim a preprint is peer-reviewed.\n- Respect arXiv 3-second polite delay (enforced in paper_search_lib).\n- Do not create files inside the skill directory; use task-local paths.\n- Ask for clarification when the topic is missing or ambiguous.\n- When a source is unavailable, degrade gracefully and report what succeeded.\n\n## Adapter Notes\n\n- Prefer bundled scripts over ad hoc shell commands for search, download, and extraction.\n- Keep the authored source of truth in split project files, not in generated adapter bundles.\n- All results are preprints. Never claim peer-review status.\n",{"name":5,"license":15,"metadata":24,"description":7,"distribution":26,"allowed-tools":27,"compatibility":28},{"short-description":25},"Download preprints as paper\u002Fcode bundles with auto-cloned GitHub repos","public","Read Write Edit Grep Glob Bash Python","Optimized for Claude Code. Install in project .claude\u002Fskills\u002F or personal ~\u002F.claude\u002Fskills\u002F. Pure local Python with no servers, no MCP, and no API keys.","main","5889e24531cb4e925523afbc48cbe1f42b5e4cd7","skills-patches\u002Fpapers\u002F.base",0,"https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack","https:\u002F\u002Fhouseofvibe.ai","https:\u002F\u002Fgithub.com\u002FStartupBros-com","2026-11-15 05:18:00",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add StartupBros-com\u002Fhov-stack --skill papers",{"tool":43,"command":44},"bunx","bunx skills add StartupBros-com\u002Fhov-stack --skill papers",{"tool":46,"command":47},"pnpm","pnpm dlx skills add StartupBros-com\u002Fhov-stack --skill papers","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-patches\u002Fpapers\u002F.base\u002FSKILL.md",true,[51,57,61],{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},"scripts\u002Fpaper_search.py","SCRIPT","脚本",115,"https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-patches\u002Fpapers\u002F.base\u002Fscripts\u002Fpaper_search.py",{"path":58,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":59,"sourceHtmlUrl":60},"scripts\u002Fworkflow.py",14125,"https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-patches\u002Fpapers\u002F.base\u002Fscripts\u002Fworkflow.py",{"path":62,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":65,"sourceHtmlUrl":66},"VERSION","OTHER","其他",2,"https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-patches\u002Fpapers\u002F.base\u002FVERSION",[68,76,84,90,97,104],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":72,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M77XGSFPK8Q1XBHYTC1TF2MV","review-adversarial-first","Operator-invoked adversarial review: 3 refute-by-default lenses plus a deterministic judge against a PR, branch, or path, then a decision on whether \u002Fce-code-review's topic fan-out runs. Off the default merge path since the 2026-10-17 merge-and-end ruling; run only when asked for by name.","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-stack\u002Freview-adversarial-first\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-stack\u002Freview-adversarial-first\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},"01M77XGNSF0ARDTK9P3EF2S4E4","memory-mine","Run one deterministic, budgeted directed-capture batch for an explicit Claude Code project store. Use manual workspace mode only when cross-agent cass sessions are intentionally in scope.","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-stack\u002Fmemory-mine\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-stack\u002Fmemory-mine\u002FSKILL.md",[],[],{"skillNo":85,"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":86,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M77XGHB8NDSNVSDWFCN1WJBY","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-patches\u002Fpapers\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fskills-patches\u002Fpapers\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M77XG4P5EM9PM8NGRZSJV4CB","wsl-cdp","Drive the user's real, logged-in Windows browser (Chrome, Edge, or Brave) from a Claude Code session running inside WSL2, over the Chrome DevTools Protocol - read authenticated dashboards and consoles, take real screenshots, open and read pages with the user's sessions. Use when a task needs the user's existing web logins and the session runs in WSL2, where the Claude in Chrome extension cannot work. Not for native-Windows or macOS Claude Code (use the official Claude in Chrome extension), not for anonymous scraping or CI (use headless Playwright inside WSL), not for getting past login walls (those are handed to the human).","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fplugins\u002Fwsl-cdp\u002Fskills\u002Fwsl-cdp\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M77XFZE48M4SQ2Y7A3GFY4RD","token-eater","[BETA] Spend a subscription's idle, about-to-expire credits on safe, machine-verified cleanup. You pick the service (Grok \u002F Codex \u002F Claude) and a token-heavy maintenance skill (de-monolithize, de-slop, simplify, dead-code); that service runs the whole skill -> gate -> self-review -> draft-PR loop on its own credits, in an isolated worktree. token-eater verifies the gate and never merges. Manual invocation only during beta.","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fplugins\u002Ftoken-eater\u002Fskills\u002Ftoken-eater\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":19},"01M77XFWRYH97KGTKCF8DF7PQR","tune","Find and fix real defects in a skill, AGENTS.md, or CLAUDE.md, then prove the description still routes. Runs the bundled token-spending runner, so it starts only on the human's explicit instruction.","https:\u002F\u002Fgithub.com\u002FStartupBros-com\u002Fhov-stack\u002Fblob\u002Fmain\u002Fplugins\u002Fskill-tuner\u002Fskills\u002Ftune\u002FSKILL.md",[],[]]