[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6M0F6DR99Y7V74A061AV47P":3,"skill-same:01M6M0F6DR99Y7V74A061AV47P":119},{"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":30,"publishedAt":58,"linkOnly":59,"markdownContent":60,"markdownBody":61,"frontMatter":62,"refName":68,"commitSha":69,"skillDirectory":70,"forksCount":13,"repositoryHtmlUrl":71,"repositoryHomepageUrl":6,"ownerHtmlUrl":72,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":73,"install":74,"resources":87},"01M6M0F6DR99Y7V74A061AV47P","research-lookup",null,"Look up current research information using parallel-cli search (primary, fast web search), the Parallel Chat API (deep research), or Perplexity sonar-pro-search (academic paper searches). Automatically routes queries to the best backend. Use for finding papers, gathering research data, and verifying scientific information.","endman100\u002Fskill-router-nexus","endman100","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F31997019?v=4","01M6KPDX17BYDWW4RRB0TNGGMX","01M6KPDX64Y2QBEWVFG5W5PWPF",0,"MIT","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002FSKILL.md",[17,22,25],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000010","后端与接口","SCENARIO","使用场景",{"categoryNo":23,"categoryName":24,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000021","资料检索与研究",{"categoryNo":26,"categoryName":27,"taxonomyType":28,"taxonomyTypeDesc":29},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57],"research","lookup","research lookup","for","cli","current research","sonar","papers","information","scientific","deep","chat","queries","up","backend","use","edit","write","read","openrouter","parallel","web search","Bash","api","perplexity","deep research","search","2026-11-10 10:12:30",false,"---\nname: research-lookup\ndescription: Look up current research information using parallel-cli search (primary, fast web search), the Parallel Chat API (deep research), or Perplexity sonar-pro-search (academic paper searches). Automatically routes queries to the best backend. Use for finding papers, gathering research data, and verifying scientific information.\nallowed-tools: Read Write Edit Bash\nlicense: MIT license\ncompatibility: parallel-cli required (primary); PARALLEL_API_KEY and OPENROUTER_API_KEY optional for deep\u002Facademic backends\nmetadata:\n    skill-author: K-Dense Inc.\n---\n\n# Research Information Lookup\n\n## Overview\n\nThis skill provides real-time research information lookup with **intelligent backend routing**:\n\n- **parallel-cli search** (parallel-web skill): **Primary and default backend** for all research queries. Fast, cost-effective web search with academic source prioritization. Uses `parallel-cli search` with `--include-domains` for scholarly sources.\n- **Parallel Chat API** (`core` model): Secondary backend for complex, multi-source deep research requiring extended synthesis (60s-5min latency). Use only when explicitly needed.\n- **Perplexity sonar-pro-search** (via OpenRouter): Used only for academic-specific paper searches where scholarly database access is critical.\n\nThe skill automatically detects query type and routes to the optimal backend.\n\n## When to Use This Skill\n\nUse this skill when you need:\n\n- **Current Research Information**: Latest studies, papers, and findings\n- **Literature Verification**: Check facts, statistics, or claims against current research\n- **Background Research**: Gather context and supporting evidence for scientific writing\n- **Citation Sources**: Find relevant papers and studies to cite\n- **Technical Documentation**: Look up specifications, protocols, or methodologies\n- **Market\u002FIndustry Data**: Current statistics, trends, competitive intelligence\n- **Recent Developments**: Emerging trends, breakthroughs, announcements\n\n## Visual Enhancement with Scientific Schematics\n\n**When creating documents with this skill, always consider adding scientific diagrams and schematics to enhance visual communication.**\n\nIf your document does not already contain schematics or diagrams:\n- Use the **scientific-schematics** skill to generate AI-powered publication-quality diagrams\n- Simply describe your desired diagram in natural language\n\n```bash\npython scripts\u002Fgenerate_schematic.py \"your diagram description\" -o figures\u002Foutput.png\n```\n\n---\n\n## Automatic Backend Selection\n\nThe skill automatically routes queries to the best backend based on content:\n\n### Routing Logic\n\n```\nQuery arrives\n    |\n    +-- Contains academic keywords? (papers, DOI, journal, peer-reviewed, etc.)\n    |       YES --> Perplexity sonar-pro-search (academic search mode)\n    |\n    +-- Needs deep multi-source synthesis? (user says \"deep research\", \"exhaustive\")\n    |       YES --> Parallel Chat API (core model, 60s-5min)\n    |\n    +-- Everything else (general research, market data, technical info, analysis)\n            --> parallel-cli search (fast, default)\n```\n\n### Default: parallel-cli search (parallel-web skill)\n\n**Primary backend for all standard research queries.** Fast, cost-effective, and supports academic source prioritization.\n\nFor scientific\u002Ftechnical queries, run two searches to ensure academic coverage:\n\n```bash\n# 1. Academic-focused search\nparallel-cli search \"your research query\" -q \"keyword1\" -q \"keyword2\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"scholar.google.com,arxiv.org,pubmed.ncbi.nlm.nih.gov,semanticscholar.org,biorxiv.org,medrxiv.org,ncbi.nlm.nih.gov,nature.com,science.org,ieee.org,acm.org,springer.com,wiley.com,cell.com,pnas.org,nih.gov\" \\\n  -o sources\u002Fresearch_\u003Ctopic>-academic.json\n\n# 2. General search (catches non-academic sources)\nparallel-cli search \"your research query\" -q \"keyword1\" -q \"keyword2\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_\u003Ctopic>-general.json\n```\n\nOptions:\n- `--after-date YYYY-MM-DD` for time-sensitive queries\n- `--include-domains domain1.com,domain2.com` to limit to specific sources\n\nMerge results, leading with academic sources. For non-scientific queries, a single general search is sufficient.\n\nAll other queries route here by default, including:\n\n- General research questions\n- Market and industry analysis\n- Technical information and documentation\n- Current events and recent developments\n- Comparative analysis\n- Statistical data retrieval\n- Fact-checking and verification\n\n### Academic Keywords (Routes to Perplexity)\n\nQueries containing these terms are routed to Perplexity for academic-focused search:\n\n- Paper finding: `find papers`, `find articles`, `research papers on`, `published studies`\n- Citations: `cite`, `citation`, `doi`, `pubmed`, `pmid`\n- Academic sources: `peer-reviewed`, `journal article`, `scholarly`, `arxiv`, `preprint`\n- Review types: `systematic review`, `meta-analysis`, `literature search`\n- Paper quality: `foundational papers`, `seminal papers`, `landmark papers`, `highly cited`\n\n### Deep Research (Routes to Parallel Chat API)\n\nOnly used when the user explicitly requests deep, exhaustive, or comprehensive research. Much slower and more expensive than parallel-cli search.\n\n### Manual Override\n\nYou can force a specific backend:\n\n```bash\n# Force parallel-cli search (fast web search)\nparallel-cli search \"your query\" -q \"keyword\" --json --max-results 10 -o sources\u002Fresearch_\u003Ctopic>.json\n\n# Force Parallel Deep Research (slow, exhaustive)\npython research_lookup.py \"your query\" --force-backend parallel\n\n# Force Perplexity academic search\npython research_lookup.py \"your query\" --force-backend perplexity\n```\n\n---\n\n## Core Capabilities\n\n### 1. General Research Queries (parallel-cli search — DEFAULT)\n\n**Primary backend.** Fast, cost-effective web search with academic source prioritization via the parallel-web skill.\n\n```\nQuery Examples:\n- \"Recent advances in CRISPR gene editing 2025\"\n- \"Compare mRNA vaccines vs traditional vaccines for cancer treatment\"\n- \"AI adoption in healthcare industry statistics\"\n- \"Global renewable energy market trends and projections\"\n- \"Explain the mechanism underlying gut microbiome and depression\"\n```\n\n```bash\n# Example: research on CRISPR advances\nparallel-cli search \"Recent advances in CRISPR gene editing 2025\" \\\n  -q \"CRISPR\" -q \"gene editing\" -q \"2025\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"scholar.google.com,arxiv.org,pubmed.ncbi.nlm.nih.gov,nature.com,science.org,cell.com,pnas.org,nih.gov\" \\\n  -o sources\u002Fresearch_crispr_advances-academic.json\n\nparallel-cli search \"Recent advances in CRISPR gene editing 2025\" \\\n  -q \"CRISPR\" -q \"gene editing\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_crispr_advances-general.json\n```\n\n**Response includes:**\n- Synthesized findings with inline citations from search results\n- Academic sources prioritized (peer-reviewed, preprints)\n- Specific facts, numbers, and dates\n- Sources section listing all referenced URLs grouped by type\n\n### 2. Academic Paper Search (Perplexity sonar-pro-search)\n\n**Used for academic-specific queries.** Prioritizes scholarly databases and peer-reviewed sources. Use when queries specifically ask for papers, citations, or DOIs.\n\n```\nQuery Examples:\n- \"Find papers on transformer attention mechanisms in NeurIPS 2024\"\n- \"Foundational papers on quantum error correction\"\n- \"Systematic review of immunotherapy in non-small cell lung cancer\"\n- \"Cite the original BERT paper and its most influential follow-ups\"\n- \"Published studies on CRISPR off-target effects in clinical trials\"\n```\n\n**Response includes:**\n- Summary of key findings from academic literature\n- 5-8 high-quality citations with authors, titles, journals, years, DOIs\n- Citation counts and venue tier indicators\n- Key statistics and methodology highlights\n- Research gaps and future directions\n\n### 3. Deep Research (Parallel Chat API — on request only)\n\n**Used only when user explicitly requests deep\u002Fexhaustive research.** Provides comprehensive, multi-source synthesis via the Chat API (`core` model). 60s-5min latency.\n\n```\nQuery Examples:\n- \"Deep research on the current state of quantum computing error correction\"\n- \"Exhaustive analysis of mRNA vaccine platforms for cancer immunotherapy\"\n```\n\n### 4. Technical and Methodological Information\n\nUse parallel-cli search (default) for quick lookups:\n\n```bash\nparallel-cli search \"Western blot protocol for protein detection\" \\\n  -q \"western blot\" -q \"protocol\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_western_blot.json\n```\n\n### 5. Statistical and Market Data\n\nUse parallel-cli search (default) for current data:\n\n```bash\nparallel-cli search \"Global AI market size and growth projections 2025\" \\\n  -q \"AI market\" -q \"statistics\" -q \"growth\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --after-date 2024-01-01 \\\n  -o sources\u002Fresearch_ai_market.json\n```\n\n---\n\n## Paper Quality and Popularity Prioritization\n\n**CRITICAL**: When searching for papers, ALWAYS prioritize high-quality, influential papers.\n\n### Citation-Based Ranking\n\n| Paper Age | Citation Threshold | Classification |\n|-----------|-------------------|----------------|\n| 0-3 years | 20+ citations | Noteworthy |\n| 0-3 years | 100+ citations | Highly Influential |\n| 3-7 years | 100+ citations | Significant |\n| 3-7 years | 500+ citations | Landmark Paper |\n| 7+ years | 500+ citations | Seminal Work |\n| 7+ years | 1000+ citations | Foundational |\n\n### Venue Quality Tiers\n\n**Tier 1 - Premier Venues** (Always prefer):\n- **General Science**: Nature, Science, Cell, PNAS\n- **Medicine**: NEJM, Lancet, JAMA, BMJ\n- **Field-Specific**: Nature Medicine, Nature Biotechnology, Nature Methods\n- **Top CS\u002FAI**: NeurIPS, ICML, ICLR, ACL, CVPR\n\n**Tier 2 - High-Impact Specialized** (Strong preference):\n- Journals with Impact Factor > 10\n- Top conferences in subfields (EMNLP, NAACL, ECCV, MICCAI)\n\n**Tier 3 - Respected Specialized** (Include when relevant):\n- Journals with Impact Factor 5-10\n\n---\n\n## Technical Integration\n\n### Prerequisites\n\n```bash\n# Primary backend (parallel-cli) - REQUIRED\n# Install parallel-cli if not already available:\ncurl -fsSL https:\u002F\u002Fparallel.ai\u002Finstall.sh | bash\n# Or: uv tool install \"parallel-web-tools[cli]\"\n\n# Authenticate:\nparallel-cli auth\n# Or: export PARALLEL_API_KEY=\"your_parallel_api_key\"\n```\n\n### Environment Variables\n\n```bash\n# Primary backend (parallel-cli search) - REQUIRED\nexport PARALLEL_API_KEY=\"your_parallel_api_key\"\n\n# Deep research backend (Parallel Chat API) - optional, for deep research only\n# Uses the same PARALLEL_API_KEY\n\n# Academic search backend (Perplexity) - optional, for academic paper queries\nexport OPENROUTER_API_KEY=\"your_openrouter_api_key\"\n```\n\n### API Specifications\n\n**parallel-cli search (PRIMARY):**\n- Command: `parallel-cli search` with `--json` output\n- Latency: 2-10 seconds (fast)\n- Output: JSON with title, URL, publish_date, excerpts\n- Academic domains: Use `--include-domains` for scholarly sources\n- Saves results: `-o filename.json` for follow-up and reproducibility\n\n**Parallel Chat API (deep research only):**\n- Endpoint: `https:\u002F\u002Fapi.parallel.ai` (OpenAI SDK compatible)\n- Model: `core` (60s-5min latency, complex multi-source synthesis)\n- Output: Markdown text with inline citations\n- Citations: Research basis with URLs, reasoning, and confidence levels\n- Rate limits: 300 req\u002Fmin\n- Python package: `openai`\n\n**Perplexity sonar-pro-search (academic only):**\n- Model: `perplexity\u002Fsonar-pro-search` (via OpenRouter)\n- Search mode: Academic (prioritizes peer-reviewed sources)\n- Search context: High (comprehensive research)\n- Response time: 5-15 seconds\n\n### Command-Line Usage\n\n```bash\n# Fast web search via parallel-cli (DEFAULT — recommended) — ALWAYS save to sources\u002F\nparallel-cli search \"your query\" -q \"keyword1\" -q \"keyword2\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_\u003Ctopic>.json\n\n# Academic-focused search via parallel-cli — ALWAYS save to sources\u002F\nparallel-cli search \"your query\" -q \"keyword1\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"scholar.google.com,arxiv.org,pubmed.ncbi.nlm.nih.gov,semanticscholar.org,biorxiv.org,medrxiv.org,nature.com,science.org,cell.com,pnas.org,nih.gov\" \\\n  -o sources\u002Fresearch_\u003Ctopic>-academic.json\n\n# Time-sensitive search via parallel-cli\nparallel-cli search \"your query\" -q \"keyword\" \\\n  --json --max-results 10 --after-date 2024-01-01 \\\n  -o sources\u002Fresearch_\u003Ctopic>.json\n\n# Extract full content from a specific URL (use parallel-web extract)\nparallel-cli extract \"https:\u002F\u002Fexample.com\u002Fpaper\" --json\n\n# Force Parallel Deep Research (slow, exhaustive) — via research_lookup.py\npython research_lookup.py \"your query\" --force-backend parallel -o sources\u002Fresearch_\u003Ctopic>.md\n\n# Force Perplexity academic search — via research_lookup.py\npython research_lookup.py \"your query\" --force-backend perplexity -o sources\u002Fpapers_\u003Ctopic>.md\n\n# Auto-routed via research_lookup.py (legacy) — ALWAYS save to sources\u002F\npython research_lookup.py \"your query\" -o sources\u002Fresearch_YYYYMMDD_HHMMSS_\u003Ctopic>.md\n\n# Batch queries via research_lookup.py — ALWAYS save to sources\u002F\npython research_lookup.py --batch \"query 1\" \"query 2\" \"query 3\" -o sources\u002Fbatch_research_\u003Ctopic>.md\n```\n\n---\n\n## MANDATORY: Save All Results to Sources Folder\n\n**Every research-lookup result MUST be saved to the project's `sources\u002F` folder.**\n\nThis is non-negotiable. Research results are expensive to obtain and critical for reproducibility.\n\n### Saving Rules\n\n| Backend | `-o` Flag Target | Filename Pattern |\n|---------|-----------------|------------------|\n| parallel-cli search (default) | `sources\u002Fresearch_\u003Ctopic>.json` | `research_\u003Cbrief_topic>.json` or `research_\u003Cbrief_topic>-academic.json` |\n| Parallel Deep Research | `sources\u002Fresearch_\u003Ctopic>.md` | `research_YYYYMMDD_HHMMSS_\u003Cbrief_topic>.md` |\n| Perplexity (academic) | `sources\u002Fpapers_\u003Ctopic>.md` | `papers_YYYYMMDD_HHMMSS_\u003Cbrief_topic>.md` |\n| Batch queries | `sources\u002Fbatch_\u003Ctopic>.md` | `batch_research_YYYYMMDD_HHMMSS_\u003Cbrief_topic>.md` |\n\n### How to Save\n\n**CRITICAL: Every search MUST save results to the `sources\u002F` folder using the `-o` flag.**\n\n**CRITICAL: Saved files MUST preserve all citations, source URLs, and DOIs.**\n\n```bash\n# parallel-cli search (DEFAULT) — save JSON to sources\u002F\nparallel-cli search \"Recent advances in CRISPR gene editing 2025\" \\\n  -q \"CRISPR\" -q \"gene editing\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"scholar.google.com,arxiv.org,pubmed.ncbi.nlm.nih.gov,nature.com,science.org,cell.com,pnas.org,nih.gov\" \\\n  -o sources\u002Fresearch_crispr_advances-academic.json\n\nparallel-cli search \"Recent advances in CRISPR gene editing 2025\" \\\n  -q \"CRISPR\" -q \"gene editing\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_crispr_advances-general.json\n\n# Academic paper search via Perplexity — save to sources\u002F\npython research_lookup.py \"Find papers on transformer attention mechanisms in NeurIPS 2024\" \\\n  -o sources\u002Fpapers_20250217_143500_transformer_attention.md\n\n# Deep research via Parallel Chat API — save to sources\u002F\npython research_lookup.py \"AI regulation landscape\" --force-backend parallel \\\n  -o sources\u002Fresearch_20250217_144000_ai_regulation.md\n\n# Batch queries — save to sources\u002F\npython research_lookup.py --batch \"mRNA vaccines efficacy\" \"mRNA vaccines safety\" \\\n  -o sources\u002Fbatch_research_20250217_144500_mrna_vaccines.md\n```\n\n### Citation Preservation in Saved Files\n\nEach output format preserves citations differently:\n\n| Format | Citations Included | When to Use |\n|--------|-------------------|-------------|\n| parallel-cli JSON (default) | Full result objects: `title`, `url`, `publish_date`, `excerpts` | Standard use — structured, parseable, fast |\n| Text (research_lookup.py) | `Sources (N):` section with `[title] (date) + URL` + `Additional References (N):` with DOIs and academic URLs | Deep research \u002F Perplexity — human-readable |\n| JSON (`--json` via research_lookup.py) | Full citation objects: `url`, `title`, `date`, `snippet`, `doi`, `type` | When you need maximum citation metadata from deep research |\n\n**For parallel-cli search**, saved JSON files include: full search results with title, URL, publish date, and content excerpts for each result.\n**For Parallel Chat API backend**, saved files include: research report + Sources list (title, URL) + Additional References (DOIs, academic URLs).\n**For Perplexity backend**, saved files include: academic summary + Sources list (title, date, URL, snippet) + Additional References (DOIs, academic URLs).\n\n**Use `--json` when you need to:**\n- Parse citation metadata programmatically\n- Preserve full DOI and URL data for BibTeX generation\n- Maintain the structured citation objects for cross-referencing\n\n### Why Save Everything\n\n1. **Reproducibility**: Every citation and claim can be traced back to its raw research source\n2. **Context Window Recovery**: If context is compacted, saved results can be re-read without re-querying\n3. **Audit Trail**: The `sources\u002F` folder documents exactly how all research information was gathered\n4. **Reuse Across Sections**: Multiple sections can reference the same saved research without duplicate queries\n5. **Cost Efficiency**: Check `sources\u002F` for existing results before making new API calls\n6. **Peer Review Support**: Reviewers can verify the research backing every citation\n\n### Before Making a New Query, Check Sources First\n\nBefore calling `research_lookup.py`, check if a relevant result already exists:\n\n```bash\nls sources\u002F  # Check existing saved results\n```\n\nIf a prior lookup covers the same topic, re-read the saved file instead of making a new API call.\n\n### Logging\n\nWhen saving research results, always log:\n\n```\n[HH:MM:SS] SAVED: Research lookup to sources\u002Fresearch_20250217_143000_crispr_advances.md (3,800 words, 8 citations)\n[HH:MM:SS] SAVED: Paper search to sources\u002Fpapers_20250217_143500_transformer_attention.md (6 papers found)\n```\n\n---\n\n## Integration with Scientific Writing\n\nThis skill enhances scientific writing by providing:\n\n1. **Literature Review Support**: Gather current research for introduction and discussion — **save to `sources\u002F`**\n2. **Methods Validation**: Verify protocols against current standards — **save to `sources\u002F`**\n3. **Results Contextualization**: Compare findings with recent similar studies — **save to `sources\u002F`**\n4. **Discussion Enhancement**: Support arguments with latest evidence — **save to `sources\u002F`**\n5. **Citation Management**: Provide properly formatted citations — **save to `sources\u002F`**\n\n## Complementary Tools\n\n| Task | Tool |\n|------|------|\n| General web search (fast) | `parallel-cli search` (built into this skill) |\n| Academic-focused web search | `parallel-cli search --include-domains` (built into this skill) |\n| URL content extraction | `parallel-cli extract` (parallel-web skill) |\n| Deep research (exhaustive) | `research-lookup` via Parallel Chat API or `parallel-web` deep research |\n| Academic paper search | `research-lookup` (auto-routes to Perplexity) |\n| Google Scholar search | `citation-management` skill |\n| PubMed search | `citation-management` skill |\n| DOI to BibTeX | `citation-management` skill |\n| Metadata verification | `parallel-cli extract` (parallel-web skill) |\n\n---\n\n## Error Handling and Limitations\n\n**Known Limitations:**\n- parallel-cli search: Requires `parallel-cli` to be installed and authenticated\n- Parallel Chat API (core model): Complex queries may take up to 5 minutes\n- Perplexity: Information cutoff, may not access full text behind paywalls\n- All backends: Cannot access proprietary or restricted databases\n\n**Fallback Behavior:**\n- If `parallel-cli` is not found, install with `curl -fsSL https:\u002F\u002Fparallel.ai\u002Finstall.sh | bash` or `uv tool install \"parallel-web-tools[cli]\"`\n- If parallel-cli search returns insufficient results, fall back to Perplexity or Parallel Chat API\n- If the selected backend's API key is missing, tries the other backend\n- If all backends fail, returns structured error response\n- Rephrase queries for better results if initial response is insufficient\n\n---\n\n## Usage Examples\n\n### Example 1: General Research (Routes to parallel-cli search)\n\n**Query**: \"Recent advances in transformer attention mechanisms 2025\"\n\n**Backend**: parallel-cli search (default, fast)\n\n**Commands**:\n```bash\nparallel-cli search \"Recent advances in transformer attention mechanisms 2025\" \\\n  -q \"transformer\" -q \"attention\" -q \"2025\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"arxiv.org,semanticscholar.org,nature.com,science.org,ieee.org,acm.org\" \\\n  -o sources\u002Fresearch_transformer_attention-academic.json\n\nparallel-cli search \"Recent advances in transformer attention mechanisms 2025\" \\\n  -q \"transformer\" -q \"attention\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_transformer_attention-general.json\n```\n\n**Response**: Synthesized findings with inline citations from academic and general sources, covering recent papers, key innovations, and performance benchmarks.\n\n### Example 2: Academic Paper Search (Routes to Perplexity)\n\n**Query**: \"Find papers on CRISPR off-target effects in clinical trials\"\n\n**Backend**: Perplexity sonar-pro-search (academic mode)\n\n**Response**: Curated list of 5-8 high-impact papers with full citations, DOIs, citation counts, and venue tier indicators.\n\n### Example 3: Comparative Analysis (Routes to parallel-cli search)\n\n**Query**: \"Compare and contrast mRNA vaccines vs traditional vaccines for cancer treatment\"\n\n**Backend**: parallel-cli search (default, fast)\n\n**Response**: Synthesized comparison from multiple web sources with inline citations, structured analysis, and evidence quality notes.\n\n### Example 4: Market Data (Routes to parallel-cli search)\n\n**Query**: \"Global AI adoption in healthcare statistics 2025\"\n\n**Backend**: parallel-cli search (default, fast)\n\n```bash\nparallel-cli search \"Global AI adoption in healthcare statistics 2025\" \\\n  -q \"AI healthcare\" -q \"adoption statistics\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --after-date 2024-01-01 \\\n  -o sources\u002Fresearch_ai_healthcare_adoption.json\n```\n\n**Response**: Current market data, adoption rates, growth projections, and regional analysis with source citations.\n\n---\n\n## Summary\n\nThis skill serves as the primary research interface with intelligent tri-backend routing:\n\n- **parallel-cli search** (default): Fast, cost-effective web search with academic source prioritization via the parallel-web skill\n- **Parallel Chat API** (`core` model): Deep, exhaustive multi-source synthesis (on explicit request only)\n- **Perplexity sonar-pro-search**: Academic-specific paper searches only\n- **Automatic routing**: Detects query type and routes to the optimal backend\n- **Manual override**: Force any backend when needed\n- **Academic prioritization**: Two-search pattern ensures scholarly sources surface for scientific queries\n","# Research Information Lookup\n\n## Overview\n\nThis skill provides real-time research information lookup with **intelligent backend routing**:\n\n- **parallel-cli search** (parallel-web skill): **Primary and default backend** for all research queries. Fast, cost-effective web search with academic source prioritization. Uses `parallel-cli search` with `--include-domains` for scholarly sources.\n- **Parallel Chat API** (`core` model): Secondary backend for complex, multi-source deep research requiring extended synthesis (60s-5min latency). Use only when explicitly needed.\n- **Perplexity sonar-pro-search** (via OpenRouter): Used only for academic-specific paper searches where scholarly database access is critical.\n\nThe skill automatically detects query type and routes to the optimal backend.\n\n## When to Use This Skill\n\nUse this skill when you need:\n\n- **Current Research Information**: Latest studies, papers, and findings\n- **Literature Verification**: Check facts, statistics, or claims against current research\n- **Background Research**: Gather context and supporting evidence for scientific writing\n- **Citation Sources**: Find relevant papers and studies to cite\n- **Technical Documentation**: Look up specifications, protocols, or methodologies\n- **Market\u002FIndustry Data**: Current statistics, trends, competitive intelligence\n- **Recent Developments**: Emerging trends, breakthroughs, announcements\n\n## Visual Enhancement with Scientific Schematics\n\n**When creating documents with this skill, always consider adding scientific diagrams and schematics to enhance visual communication.**\n\nIf your document does not already contain schematics or diagrams:\n- Use the **scientific-schematics** skill to generate AI-powered publication-quality diagrams\n- Simply describe your desired diagram in natural language\n\n```bash\npython scripts\u002Fgenerate_schematic.py \"your diagram description\" -o figures\u002Foutput.png\n```\n\n---\n\n## Automatic Backend Selection\n\nThe skill automatically routes queries to the best backend based on content:\n\n### Routing Logic\n\n```\nQuery arrives\n    |\n    +-- Contains academic keywords? (papers, DOI, journal, peer-reviewed, etc.)\n    |       YES --> Perplexity sonar-pro-search (academic search mode)\n    |\n    +-- Needs deep multi-source synthesis? (user says \"deep research\", \"exhaustive\")\n    |       YES --> Parallel Chat API (core model, 60s-5min)\n    |\n    +-- Everything else (general research, market data, technical info, analysis)\n            --> parallel-cli search (fast, default)\n```\n\n### Default: parallel-cli search (parallel-web skill)\n\n**Primary backend for all standard research queries.** Fast, cost-effective, and supports academic source prioritization.\n\nFor scientific\u002Ftechnical queries, run two searches to ensure academic coverage:\n\n```bash\n# 1. Academic-focused search\nparallel-cli search \"your research query\" -q \"keyword1\" -q \"keyword2\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"scholar.google.com,arxiv.org,pubmed.ncbi.nlm.nih.gov,semanticscholar.org,biorxiv.org,medrxiv.org,ncbi.nlm.nih.gov,nature.com,science.org,ieee.org,acm.org,springer.com,wiley.com,cell.com,pnas.org,nih.gov\" \\\n  -o sources\u002Fresearch_\u003Ctopic>-academic.json\n\n# 2. General search (catches non-academic sources)\nparallel-cli search \"your research query\" -q \"keyword1\" -q \"keyword2\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_\u003Ctopic>-general.json\n```\n\nOptions:\n- `--after-date YYYY-MM-DD` for time-sensitive queries\n- `--include-domains domain1.com,domain2.com` to limit to specific sources\n\nMerge results, leading with academic sources. For non-scientific queries, a single general search is sufficient.\n\nAll other queries route here by default, including:\n\n- General research questions\n- Market and industry analysis\n- Technical information and documentation\n- Current events and recent developments\n- Comparative analysis\n- Statistical data retrieval\n- Fact-checking and verification\n\n### Academic Keywords (Routes to Perplexity)\n\nQueries containing these terms are routed to Perplexity for academic-focused search:\n\n- Paper finding: `find papers`, `find articles`, `research papers on`, `published studies`\n- Citations: `cite`, `citation`, `doi`, `pubmed`, `pmid`\n- Academic sources: `peer-reviewed`, `journal article`, `scholarly`, `arxiv`, `preprint`\n- Review types: `systematic review`, `meta-analysis`, `literature search`\n- Paper quality: `foundational papers`, `seminal papers`, `landmark papers`, `highly cited`\n\n### Deep Research (Routes to Parallel Chat API)\n\nOnly used when the user explicitly requests deep, exhaustive, or comprehensive research. Much slower and more expensive than parallel-cli search.\n\n### Manual Override\n\nYou can force a specific backend:\n\n```bash\n# Force parallel-cli search (fast web search)\nparallel-cli search \"your query\" -q \"keyword\" --json --max-results 10 -o sources\u002Fresearch_\u003Ctopic>.json\n\n# Force Parallel Deep Research (slow, exhaustive)\npython research_lookup.py \"your query\" --force-backend parallel\n\n# Force Perplexity academic search\npython research_lookup.py \"your query\" --force-backend perplexity\n```\n\n---\n\n## Core Capabilities\n\n### 1. General Research Queries (parallel-cli search — DEFAULT)\n\n**Primary backend.** Fast, cost-effective web search with academic source prioritization via the parallel-web skill.\n\n```\nQuery Examples:\n- \"Recent advances in CRISPR gene editing 2025\"\n- \"Compare mRNA vaccines vs traditional vaccines for cancer treatment\"\n- \"AI adoption in healthcare industry statistics\"\n- \"Global renewable energy market trends and projections\"\n- \"Explain the mechanism underlying gut microbiome and depression\"\n```\n\n```bash\n# Example: research on CRISPR advances\nparallel-cli search \"Recent advances in CRISPR gene editing 2025\" \\\n  -q \"CRISPR\" -q \"gene editing\" -q \"2025\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"scholar.google.com,arxiv.org,pubmed.ncbi.nlm.nih.gov,nature.com,science.org,cell.com,pnas.org,nih.gov\" \\\n  -o sources\u002Fresearch_crispr_advances-academic.json\n\nparallel-cli search \"Recent advances in CRISPR gene editing 2025\" \\\n  -q \"CRISPR\" -q \"gene editing\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_crispr_advances-general.json\n```\n\n**Response includes:**\n- Synthesized findings with inline citations from search results\n- Academic sources prioritized (peer-reviewed, preprints)\n- Specific facts, numbers, and dates\n- Sources section listing all referenced URLs grouped by type\n\n### 2. Academic Paper Search (Perplexity sonar-pro-search)\n\n**Used for academic-specific queries.** Prioritizes scholarly databases and peer-reviewed sources. Use when queries specifically ask for papers, citations, or DOIs.\n\n```\nQuery Examples:\n- \"Find papers on transformer attention mechanisms in NeurIPS 2024\"\n- \"Foundational papers on quantum error correction\"\n- \"Systematic review of immunotherapy in non-small cell lung cancer\"\n- \"Cite the original BERT paper and its most influential follow-ups\"\n- \"Published studies on CRISPR off-target effects in clinical trials\"\n```\n\n**Response includes:**\n- Summary of key findings from academic literature\n- 5-8 high-quality citations with authors, titles, journals, years, DOIs\n- Citation counts and venue tier indicators\n- Key statistics and methodology highlights\n- Research gaps and future directions\n\n### 3. Deep Research (Parallel Chat API — on request only)\n\n**Used only when user explicitly requests deep\u002Fexhaustive research.** Provides comprehensive, multi-source synthesis via the Chat API (`core` model). 60s-5min latency.\n\n```\nQuery Examples:\n- \"Deep research on the current state of quantum computing error correction\"\n- \"Exhaustive analysis of mRNA vaccine platforms for cancer immunotherapy\"\n```\n\n### 4. Technical and Methodological Information\n\nUse parallel-cli search (default) for quick lookups:\n\n```bash\nparallel-cli search \"Western blot protocol for protein detection\" \\\n  -q \"western blot\" -q \"protocol\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_western_blot.json\n```\n\n### 5. Statistical and Market Data\n\nUse parallel-cli search (default) for current data:\n\n```bash\nparallel-cli search \"Global AI market size and growth projections 2025\" \\\n  -q \"AI market\" -q \"statistics\" -q \"growth\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --after-date 2024-01-01 \\\n  -o sources\u002Fresearch_ai_market.json\n```\n\n---\n\n## Paper Quality and Popularity Prioritization\n\n**CRITICAL**: When searching for papers, ALWAYS prioritize high-quality, influential papers.\n\n### Citation-Based Ranking\n\n| Paper Age | Citation Threshold | Classification |\n|-----------|-------------------|----------------|\n| 0-3 years | 20+ citations | Noteworthy |\n| 0-3 years | 100+ citations | Highly Influential |\n| 3-7 years | 100+ citations | Significant |\n| 3-7 years | 500+ citations | Landmark Paper |\n| 7+ years | 500+ citations | Seminal Work |\n| 7+ years | 1000+ citations | Foundational |\n\n### Venue Quality Tiers\n\n**Tier 1 - Premier Venues** (Always prefer):\n- **General Science**: Nature, Science, Cell, PNAS\n- **Medicine**: NEJM, Lancet, JAMA, BMJ\n- **Field-Specific**: Nature Medicine, Nature Biotechnology, Nature Methods\n- **Top CS\u002FAI**: NeurIPS, ICML, ICLR, ACL, CVPR\n\n**Tier 2 - High-Impact Specialized** (Strong preference):\n- Journals with Impact Factor > 10\n- Top conferences in subfields (EMNLP, NAACL, ECCV, MICCAI)\n\n**Tier 3 - Respected Specialized** (Include when relevant):\n- Journals with Impact Factor 5-10\n\n---\n\n## Technical Integration\n\n### Prerequisites\n\n```bash\n# Primary backend (parallel-cli) - REQUIRED\n# Install parallel-cli if not already available:\ncurl -fsSL https:\u002F\u002Fparallel.ai\u002Finstall.sh | bash\n# Or: uv tool install \"parallel-web-tools[cli]\"\n\n# Authenticate:\nparallel-cli auth\n# Or: export PARALLEL_API_KEY=\"your_parallel_api_key\"\n```\n\n### Environment Variables\n\n```bash\n# Primary backend (parallel-cli search) - REQUIRED\nexport PARALLEL_API_KEY=\"your_parallel_api_key\"\n\n# Deep research backend (Parallel Chat API) - optional, for deep research only\n# Uses the same PARALLEL_API_KEY\n\n# Academic search backend (Perplexity) - optional, for academic paper queries\nexport OPENROUTER_API_KEY=\"your_openrouter_api_key\"\n```\n\n### API Specifications\n\n**parallel-cli search (PRIMARY):**\n- Command: `parallel-cli search` with `--json` output\n- Latency: 2-10 seconds (fast)\n- Output: JSON with title, URL, publish_date, excerpts\n- Academic domains: Use `--include-domains` for scholarly sources\n- Saves results: `-o filename.json` for follow-up and reproducibility\n\n**Parallel Chat API (deep research only):**\n- Endpoint: `https:\u002F\u002Fapi.parallel.ai` (OpenAI SDK compatible)\n- Model: `core` (60s-5min latency, complex multi-source synthesis)\n- Output: Markdown text with inline citations\n- Citations: Research basis with URLs, reasoning, and confidence levels\n- Rate limits: 300 req\u002Fmin\n- Python package: `openai`\n\n**Perplexity sonar-pro-search (academic only):**\n- Model: `perplexity\u002Fsonar-pro-search` (via OpenRouter)\n- Search mode: Academic (prioritizes peer-reviewed sources)\n- Search context: High (comprehensive research)\n- Response time: 5-15 seconds\n\n### Command-Line Usage\n\n```bash\n# Fast web search via parallel-cli (DEFAULT — recommended) — ALWAYS save to sources\u002F\nparallel-cli search \"your query\" -q \"keyword1\" -q \"keyword2\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_\u003Ctopic>.json\n\n# Academic-focused search via parallel-cli — ALWAYS save to sources\u002F\nparallel-cli search \"your query\" -q \"keyword1\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"scholar.google.com,arxiv.org,pubmed.ncbi.nlm.nih.gov,semanticscholar.org,biorxiv.org,medrxiv.org,nature.com,science.org,cell.com,pnas.org,nih.gov\" \\\n  -o sources\u002Fresearch_\u003Ctopic>-academic.json\n\n# Time-sensitive search via parallel-cli\nparallel-cli search \"your query\" -q \"keyword\" \\\n  --json --max-results 10 --after-date 2024-01-01 \\\n  -o sources\u002Fresearch_\u003Ctopic>.json\n\n# Extract full content from a specific URL (use parallel-web extract)\nparallel-cli extract \"https:\u002F\u002Fexample.com\u002Fpaper\" --json\n\n# Force Parallel Deep Research (slow, exhaustive) — via research_lookup.py\npython research_lookup.py \"your query\" --force-backend parallel -o sources\u002Fresearch_\u003Ctopic>.md\n\n# Force Perplexity academic search — via research_lookup.py\npython research_lookup.py \"your query\" --force-backend perplexity -o sources\u002Fpapers_\u003Ctopic>.md\n\n# Auto-routed via research_lookup.py (legacy) — ALWAYS save to sources\u002F\npython research_lookup.py \"your query\" -o sources\u002Fresearch_YYYYMMDD_HHMMSS_\u003Ctopic>.md\n\n# Batch queries via research_lookup.py — ALWAYS save to sources\u002F\npython research_lookup.py --batch \"query 1\" \"query 2\" \"query 3\" -o sources\u002Fbatch_research_\u003Ctopic>.md\n```\n\n---\n\n## MANDATORY: Save All Results to Sources Folder\n\n**Every research-lookup result MUST be saved to the project's `sources\u002F` folder.**\n\nThis is non-negotiable. Research results are expensive to obtain and critical for reproducibility.\n\n### Saving Rules\n\n| Backend | `-o` Flag Target | Filename Pattern |\n|---------|-----------------|------------------|\n| parallel-cli search (default) | `sources\u002Fresearch_\u003Ctopic>.json` | `research_\u003Cbrief_topic>.json` or `research_\u003Cbrief_topic>-academic.json` |\n| Parallel Deep Research | `sources\u002Fresearch_\u003Ctopic>.md` | `research_YYYYMMDD_HHMMSS_\u003Cbrief_topic>.md` |\n| Perplexity (academic) | `sources\u002Fpapers_\u003Ctopic>.md` | `papers_YYYYMMDD_HHMMSS_\u003Cbrief_topic>.md` |\n| Batch queries | `sources\u002Fbatch_\u003Ctopic>.md` | `batch_research_YYYYMMDD_HHMMSS_\u003Cbrief_topic>.md` |\n\n### How to Save\n\n**CRITICAL: Every search MUST save results to the `sources\u002F` folder using the `-o` flag.**\n\n**CRITICAL: Saved files MUST preserve all citations, source URLs, and DOIs.**\n\n```bash\n# parallel-cli search (DEFAULT) — save JSON to sources\u002F\nparallel-cli search \"Recent advances in CRISPR gene editing 2025\" \\\n  -q \"CRISPR\" -q \"gene editing\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"scholar.google.com,arxiv.org,pubmed.ncbi.nlm.nih.gov,nature.com,science.org,cell.com,pnas.org,nih.gov\" \\\n  -o sources\u002Fresearch_crispr_advances-academic.json\n\nparallel-cli search \"Recent advances in CRISPR gene editing 2025\" \\\n  -q \"CRISPR\" -q \"gene editing\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_crispr_advances-general.json\n\n# Academic paper search via Perplexity — save to sources\u002F\npython research_lookup.py \"Find papers on transformer attention mechanisms in NeurIPS 2024\" \\\n  -o sources\u002Fpapers_20250217_143500_transformer_attention.md\n\n# Deep research via Parallel Chat API — save to sources\u002F\npython research_lookup.py \"AI regulation landscape\" --force-backend parallel \\\n  -o sources\u002Fresearch_20250217_144000_ai_regulation.md\n\n# Batch queries — save to sources\u002F\npython research_lookup.py --batch \"mRNA vaccines efficacy\" \"mRNA vaccines safety\" \\\n  -o sources\u002Fbatch_research_20250217_144500_mrna_vaccines.md\n```\n\n### Citation Preservation in Saved Files\n\nEach output format preserves citations differently:\n\n| Format | Citations Included | When to Use |\n|--------|-------------------|-------------|\n| parallel-cli JSON (default) | Full result objects: `title`, `url`, `publish_date`, `excerpts` | Standard use — structured, parseable, fast |\n| Text (research_lookup.py) | `Sources (N):` section with `[title] (date) + URL` + `Additional References (N):` with DOIs and academic URLs | Deep research \u002F Perplexity — human-readable |\n| JSON (`--json` via research_lookup.py) | Full citation objects: `url`, `title`, `date`, `snippet`, `doi`, `type` | When you need maximum citation metadata from deep research |\n\n**For parallel-cli search**, saved JSON files include: full search results with title, URL, publish date, and content excerpts for each result.\n**For Parallel Chat API backend**, saved files include: research report + Sources list (title, URL) + Additional References (DOIs, academic URLs).\n**For Perplexity backend**, saved files include: academic summary + Sources list (title, date, URL, snippet) + Additional References (DOIs, academic URLs).\n\n**Use `--json` when you need to:**\n- Parse citation metadata programmatically\n- Preserve full DOI and URL data for BibTeX generation\n- Maintain the structured citation objects for cross-referencing\n\n### Why Save Everything\n\n1. **Reproducibility**: Every citation and claim can be traced back to its raw research source\n2. **Context Window Recovery**: If context is compacted, saved results can be re-read without re-querying\n3. **Audit Trail**: The `sources\u002F` folder documents exactly how all research information was gathered\n4. **Reuse Across Sections**: Multiple sections can reference the same saved research without duplicate queries\n5. **Cost Efficiency**: Check `sources\u002F` for existing results before making new API calls\n6. **Peer Review Support**: Reviewers can verify the research backing every citation\n\n### Before Making a New Query, Check Sources First\n\nBefore calling `research_lookup.py`, check if a relevant result already exists:\n\n```bash\nls sources\u002F  # Check existing saved results\n```\n\nIf a prior lookup covers the same topic, re-read the saved file instead of making a new API call.\n\n### Logging\n\nWhen saving research results, always log:\n\n```\n[HH:MM:SS] SAVED: Research lookup to sources\u002Fresearch_20250217_143000_crispr_advances.md (3,800 words, 8 citations)\n[HH:MM:SS] SAVED: Paper search to sources\u002Fpapers_20250217_143500_transformer_attention.md (6 papers found)\n```\n\n---\n\n## Integration with Scientific Writing\n\nThis skill enhances scientific writing by providing:\n\n1. **Literature Review Support**: Gather current research for introduction and discussion — **save to `sources\u002F`**\n2. **Methods Validation**: Verify protocols against current standards — **save to `sources\u002F`**\n3. **Results Contextualization**: Compare findings with recent similar studies — **save to `sources\u002F`**\n4. **Discussion Enhancement**: Support arguments with latest evidence — **save to `sources\u002F`**\n5. **Citation Management**: Provide properly formatted citations — **save to `sources\u002F`**\n\n## Complementary Tools\n\n| Task | Tool |\n|------|------|\n| General web search (fast) | `parallel-cli search` (built into this skill) |\n| Academic-focused web search | `parallel-cli search --include-domains` (built into this skill) |\n| URL content extraction | `parallel-cli extract` (parallel-web skill) |\n| Deep research (exhaustive) | `research-lookup` via Parallel Chat API or `parallel-web` deep research |\n| Academic paper search | `research-lookup` (auto-routes to Perplexity) |\n| Google Scholar search | `citation-management` skill |\n| PubMed search | `citation-management` skill |\n| DOI to BibTeX | `citation-management` skill |\n| Metadata verification | `parallel-cli extract` (parallel-web skill) |\n\n---\n\n## Error Handling and Limitations\n\n**Known Limitations:**\n- parallel-cli search: Requires `parallel-cli` to be installed and authenticated\n- Parallel Chat API (core model): Complex queries may take up to 5 minutes\n- Perplexity: Information cutoff, may not access full text behind paywalls\n- All backends: Cannot access proprietary or restricted databases\n\n**Fallback Behavior:**\n- If `parallel-cli` is not found, install with `curl -fsSL https:\u002F\u002Fparallel.ai\u002Finstall.sh | bash` or `uv tool install \"parallel-web-tools[cli]\"`\n- If parallel-cli search returns insufficient results, fall back to Perplexity or Parallel Chat API\n- If the selected backend's API key is missing, tries the other backend\n- If all backends fail, returns structured error response\n- Rephrase queries for better results if initial response is insufficient\n\n---\n\n## Usage Examples\n\n### Example 1: General Research (Routes to parallel-cli search)\n\n**Query**: \"Recent advances in transformer attention mechanisms 2025\"\n\n**Backend**: parallel-cli search (default, fast)\n\n**Commands**:\n```bash\nparallel-cli search \"Recent advances in transformer attention mechanisms 2025\" \\\n  -q \"transformer\" -q \"attention\" -q \"2025\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --include-domains \"arxiv.org,semanticscholar.org,nature.com,science.org,ieee.org,acm.org\" \\\n  -o sources\u002Fresearch_transformer_attention-academic.json\n\nparallel-cli search \"Recent advances in transformer attention mechanisms 2025\" \\\n  -q \"transformer\" -q \"attention\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  -o sources\u002Fresearch_transformer_attention-general.json\n```\n\n**Response**: Synthesized findings with inline citations from academic and general sources, covering recent papers, key innovations, and performance benchmarks.\n\n### Example 2: Academic Paper Search (Routes to Perplexity)\n\n**Query**: \"Find papers on CRISPR off-target effects in clinical trials\"\n\n**Backend**: Perplexity sonar-pro-search (academic mode)\n\n**Response**: Curated list of 5-8 high-impact papers with full citations, DOIs, citation counts, and venue tier indicators.\n\n### Example 3: Comparative Analysis (Routes to parallel-cli search)\n\n**Query**: \"Compare and contrast mRNA vaccines vs traditional vaccines for cancer treatment\"\n\n**Backend**: parallel-cli search (default, fast)\n\n**Response**: Synthesized comparison from multiple web sources with inline citations, structured analysis, and evidence quality notes.\n\n### Example 4: Market Data (Routes to parallel-cli search)\n\n**Query**: \"Global AI adoption in healthcare statistics 2025\"\n\n**Backend**: parallel-cli search (default, fast)\n\n```bash\nparallel-cli search \"Global AI adoption in healthcare statistics 2025\" \\\n  -q \"AI healthcare\" -q \"adoption statistics\" \\\n  --json --max-results 10 --excerpt-max-chars-total 27000 \\\n  --after-date 2024-01-01 \\\n  -o sources\u002Fresearch_ai_healthcare_adoption.json\n```\n\n**Response**: Current market data, adoption rates, growth projections, and regional analysis with source citations.\n\n---\n\n## Summary\n\nThis skill serves as the primary research interface with intelligent tri-backend routing:\n\n- **parallel-cli search** (default): Fast, cost-effective web search with academic source prioritization via the parallel-web skill\n- **Parallel Chat API** (`core` model): Deep, exhaustive multi-source synthesis (on explicit request only)\n- **Perplexity sonar-pro-search**: Academic-specific paper searches only\n- **Automatic routing**: Detects query type and routes to the optimal backend\n- **Manual override**: Force any backend when needed\n- **Academic prioritization**: Two-search pattern ensures scholarly sources surface for scientific queries\n",{"name":5,"license":63,"metadata":64,"description":7,"allowed-tools":66,"compatibility":67},"MIT license",{"skill-author":65},"K-Dense Inc.","Read Write Edit Bash","parallel-cli required (primary); PARALLEL_API_KEY and OPENROUTER_API_KEY optional for deep\u002Facademic backends","main","f19b254e8975d0bfddb394686706106930d7964d","Science\u002Fresearch-and-communication\u002Fresearch-lookup","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus","https:\u002F\u002Fgithub.com\u002Fendman100","2026-11-12 18:13:37",{"source":8,"skillName":5,"commands":75,"prompt":85,"zipAvailable":86},[76,79,82],{"tool":77,"command":78},"npx","npx skills add endman100\u002Fskill-router-nexus --skill research-lookup",{"tool":80,"command":81},"bunx","bunx skills add endman100\u002Fskill-router-nexus --skill research-lookup",{"tool":83,"command":84},"pnpm","pnpm dlx skills add endman100\u002Fskill-router-nexus --skill research-lookup","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002FSKILL.md",true,[88,94,98,104,108,112,116],{"path":89,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":92,"sourceHtmlUrl":93},"examples.py","SCRIPT","脚本",5672,"https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002Fexamples.py",{"path":95,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":96,"sourceHtmlUrl":97},"lookup.py",5957,"https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002Flookup.py",{"path":99,"resourceType":100,"resourceTypeDesc":101,"sizeBytes":102,"sourceHtmlUrl":103},"README.md","REFERENCE","参考文档",5606,"https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002FREADME.md",{"path":105,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":106,"sourceHtmlUrl":107},"research_lookup.py",22642,"https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002Fresearch_lookup.py",{"path":109,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":110,"sourceHtmlUrl":111},"scripts\u002Fgenerate_schematic_ai.py",32647,"https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002Fscripts\u002Fgenerate_schematic_ai.py",{"path":113,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":114,"sourceHtmlUrl":115},"scripts\u002Fgenerate_schematic.py",5146,"https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002Fscripts\u002Fgenerate_schematic.py",{"path":117,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":106,"sourceHtmlUrl":118},"scripts\u002Fresearch_lookup.py","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FScience\u002Fresearch-and-communication\u002Fresearch-lookup\u002Fscripts\u002Fresearch_lookup.py",[120,138,172,195,204,228],{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":124,"sourceHtmlUrl":125,"categories":126,"keywords":130,"publishedAt":58},"01M6MPCT5BXVER8HQZTW3QE9AS","create-master","基于佛教经典文献，生成特定高僧大德的 AI 教学角色","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FMetaphysics-and-Spirituality\u002FMaster-skill\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FMetaphysics-and-Spirituality\u002FMaster-skill\u002FSKILL.md",[127],{"categoryNo":128,"categoryName":129,"taxonomyType":28,"taxonomyTypeDesc":29},"01JBSEEDCAT000000000000039","教师与教育",[131,132,133,134,135,136,47,48,49,137,53],"create master","master","create","生成","true","webfetch","grep",{"skillNo":139,"name":140,"nameZh":6,"description":141,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":142,"sourceHtmlUrl":143,"categories":144,"keywords":154,"publishedAt":58},"01M6M43AKE95Y64E1JRXEF5QQ3","writing-dna-skill","从至少 20 篇完整文章中蒸馏可复用的写作 DNA，分析语言、文章结构、选题逻辑、素材策略、认知框架和视觉风格，并生成 Writing-DNA.md；按该 DNA 写作时会读取全部蒸馏产物和 5 篇相关原文。用于中英文作者、账号、品牌或出版物的风格分析与一致性写作。 Distill reusable Writing DNA from at least 20 complete articles for Chinese or English authors, publications, brands, and accounts; use for language, structure, topic logic, source strategy, cognitive-frame, and visual-style analysis.","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fwriting-dna-skill\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fwriting-dna-skill\u002FSKILL.md",[145,148,151],{"categoryNo":146,"categoryName":147,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":149,"categoryName":150,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":152,"categoryName":153,"taxonomyType":28,"taxonomyTypeDesc":29},"01JBSEEDCAT000000000000036","内容创作者",[155,156,157,158,159,160,161,162,163,164,46,165,166,134,167,168,169,34,170,171],"writing dna skill","DNA","writing","strategy","english","logic","visual","structure","chinese","reusable","style","distill","analysis","frame","source","at","complete",{"skillNo":173,"name":174,"nameZh":6,"description":175,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":176,"sourceHtmlUrl":177,"categories":178,"keywords":181,"publishedAt":58},"01M6M428TTCGQ4ME3KY99VN037","stop-slop","Remove AI writing patterns from prose. Use when drafting, editing, or reviewing text to eliminate predictable AI tells.","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fstop-slop\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fstop-slop\u002FSKILL.md",[179,180],{"categoryNo":146,"categoryName":147,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":152,"categoryName":153,"taxonomyType":28,"taxonomyTypeDesc":29},[182,183,184,185,186,187,188,189,190,46,191,192,193,34,194,157],"slop","stop slop","stop","drafting","when","drafts","https","patterns","prose","remove","editing","text","content",{"skillNo":196,"name":197,"nameZh":6,"description":198,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":199,"sourceHtmlUrl":200,"categories":201,"keywords":202,"publishedAt":58},"01M6M41X0B7EBX42TK435DMGGR","shuorenhua","检查和清理中英文文本里的 AI 套路，适用于“去 AI 味”“说人话”“自然一点”“别像模板”“先标问题”这类改写和审稿需求；按场景控制力度，同时保留事实、术语、语域和责任主体。","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fshuorenhua\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fshuorenhua\u002FSKILL.md",[],[197,203],"文本",{"skillNo":205,"name":206,"nameZh":6,"description":207,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":208,"sourceHtmlUrl":209,"categories":210,"keywords":211,"publishedAt":58},"01M6M3ZD12WBEW5KXKDM4N6DFG","naming","Name products, SaaS, brands, open source projects, bots, and apps. Use when the user needs to name something, find a brand name, or pick a product name. Metaphor-driven process that produces memorable, meaningful names and avoids AI slop.","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fnaming-main\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fnaming-main\u002FSKILL.md",[],[206,212,213,214,215,216,46,217,49,169,218,219,220,221,53,222,223,224,225,186,226,227,182],"repo","websearch","product name","apps","bots","view","brand","pick","gh","projects","open","saas","open source","product","describe","ai slop",{"skillNo":229,"name":230,"nameZh":6,"description":231,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":232,"sourceHtmlUrl":233,"categories":234,"keywords":236,"publishedAt":58},"01M6M3YVXAFAAWJ7F6JMPFREYE","humanizer-zh","去除文本中的 AI 生成痕迹。适用于编辑或审阅文本，使其听起来更自然、更像人类书写。\n基于维基百科的\"AI 写作特征\"综合指南。检测并修复以下模式：夸大的象征意义、\n宣传性语言、以 -ing 结尾的肤浅分析、模糊的归因、破折号过度使用、三段式法则、\nAI 词汇、否定式排比、过多的连接性短语。","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fhumanizer-zh\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fendman100\u002Fskill-router-nexus\u002Fblob\u002Fmain\u002FWriting-and-Content\u002Fhumanizer-zh\u002FSKILL.md",[235],{"categoryNo":146,"categoryName":147,"taxonomyType":20,"taxonomyTypeDesc":21},[237,238,239,49,48,47,157,134,203,240,182,184],"humanizer","zh","humanizer zh","翻译"]