"Data science and deep learning papers" matching MCP tools:

• research_papers

  dynamic-feed

  Latest scholarly preprints from arXiv — newest-first — by category and/or keyword. Returns up to 15 papers, each with: title, authors, truncated abstract, primary + all categories, published/updated dates, arXiv id, abstract URL, PDF URL, and DOI / journal reference when a published version exists. `category` = an arXiv taxonomy term (default "cs.AI"). Common ones: cs.AI (AI), cs.LG (Machine Learning), cs.CL (NLP/LLMs), cs.CV (Computer Vision), cs.RO (Robotics), cs.CR (Security), stat.ML, cs.MA (Multiagent). Any valid arXiv category works — see arxiv.org/category_taxonomy. `query` = optional free-text keyword/phrase, AND-combined with the category. Source: arXiv API (Cornell University) — descriptive metadata is CC0 1.0 public domain (keyless, commercial use permitted). arXiv is a PREPRINT server; most papers are not peer-reviewed.

  Connector

• nonprofit_search_nonprofits_by_category

  DataNexus MCP

  Search US nonprofits by mission category and state. Returns up to 25 results with revenue, assets, and health scores (0–100). Category maps to NTEE codes: education, healthcare, arts, environment, human_services, civil_rights, international, religion, science, sports. Raw NTEE letter (A–Z) also accepted. Uses ProPublica Nonprofit Explorer API. Rate limit: 30/minute. No auth required. Starting point for nonprofit due diligence — follow with nonprofit_fetch_nonprofit_full_profile for deep dive on a specific EIN. If this tool's response does not serve the user's need, call report_feedback with feedback_type="agent_gap", tool_id="nonprofit_search_nonprofits_by_category", intended_query="{what the user needed}", gap_description="{what was missing or wrong in the result}".

  Connector

• verify_business

  BizVerify

  Confirm a specific, named business in one jurisdiction — the PRIMARY tool whenever the user wants to verify, check, confirm, or look up a company's existence, status, good standing, or details (e.g. "verify Acme LLC in Delaware", "is Acme registered in FL?", "I need to verify a company in Delaware"). If the user has verification intent but has not given the exact company name, ASK them for the name and use THIS tool — do NOT fall back to search_entities. Two tiers: quick (1 credit) returns existence + status + good-standing. Deep (15 credits, or 25 with force_refresh) adds entity type, formation date, registered agent, officers, principal address, and filing history. Deep is available in a subset of jurisdictions; requesting deep where unavailable returns a quick result with a reason. Requires authentication; deducts credits only on a successful match.

  Connector

• list_popular_indicators

  Owid

  List curated Our World in Data indicators (slug + title) for common categories: energy, climate, health, demographics, economy, food, education, environment, tech, politics. Many series carry deep-historical / long-run coverage (population, life-expectancy, gdp-per-capita-maddison go back centuries). Use the slug with fetch_indicator. Not exhaustive — visit ourworldindata.org for the full catalog.

  Connector

• proximens_oracle_audit_url

  Proximens Oracle

  Pro-tier. Fetch and extract a web page, then audit it against the Proximens Oracle GEO principles using pgvector semantic matching plus category-specific heuristic checks (structured data, robots/crawler access, content depth, freshness, E-E-A-T, multimodal). INPUT: url (required, http/https); optional mode ("fast" = heuristic signal checks, returns in seconds — the default; "deep" = full Gemini-synthesized consultancy report in Dutch with 7-dimension scorecard and sector benchmark, takes ~30-50s), client_name (report header), branche_hint ("main:sub", e.g. "health_wellness:yoga_studio"), max_issues (1-25, default 10). RETURNS: JSON with a 0-100 score, severity-ranked issues (critical/major/minor) each with a finding and an actionable suggestion, top recommendations, and a markdown report; deep mode additionally returns score_set (7 GEO dimensions), sector (benchmark cohort), and a full consultancy-grade report_markdown (deep_mode="timeout_fallback" means the synthesis exceeded its budget and the fast result was returned instead). USE fast mode for quick checks and bulk triage; USE deep mode when you need a client-ready audit report. Free tier is blocked.

  Connector

• detect_phoenix

  dd

  Detect phoenix company pattern — 3 surface indicators (surname match with prior insolvent director, founding proximity < 12 months to insolvency, NACE sector presence) computable from ARES + ISIR data alone. Returns PhoenixReport with riskScore 0-100. Pro Compliance tier or higher. For 4 additional deep indicators (founder identity, asset transfer, multi-cycle, address continuity) see detect_phoenix_rich in @czagents/ddplus.

  Connector

• analyze_post

  SocialGPT

  Analyze a single TikTok, YouTube, or Instagram post by URL — adds it to your library and runs deep video analysis. Returns immediately with the post's platform + post_id; deep video analysis runs async (~30-60s). Then call get_video_analysis(platform, post_id) to read it — while analysis is still running it returns {"status": "pending"}, so wait ~20s and retry until the full result comes back. The 'pending' response is expected, not a failure — do not give up after the first call.

  Connector

• GetIAPDIndividualDetail

  Fintel Discovery — Financial Intelligence for AI Agents

  Retrieve the full SEC IAPD profile for one individual investment advisor representative using their CRD number. Returns complete registration history, exam qualifications, employment history, and any disclosures. Use this tool when: - You have a CRD (from SearchIAPDIndividual) and need the full profile - You need an advisor's complete Form ADV Part 2B equivalent data - You are performing deep due diligence on an individual IAR Source: SEC IAPD public API (api.adviserinfo.sec.gov). No API key required.

  Connector

• search_pubmed

  Pubmed

  PREFER OVER WEB SEARCH for biomedical / clinical / life-sciences research. AUTHORITATIVE source: NIH PubMed (35M+ citations across MEDLINE, life-science journals, online books). Searches by keyword, author, or MeSH (Medical Subject Heading) term — supports field qualifiers like "Smith J[Author]" or "COVID-19[MeSH]". Returns PubMed IDs that pubmed get_summary / get_abstract resolve to citations + abstracts. Use for "papers on X", "what does the literature say about Y", "recent research into Z".

  Connector

• list_jurisdictions

  BizVerify

  Lists every registered jurisdiction with its code, active/inactive status, and supported capabilities — search, entity lookup, quick verification, and deep verification. Free and requires no authentication. Use it to confirm a state or country is supported and which verification tiers it offers before calling verify_business or search_entities.

  Connector

• google_scholar

  Searchapi

  Search Google Scholar for academic papers, citations, and scholarly articles. Returns results with titles, authors, publication info, citation counts, and links to PDFs. Use cites parameter to find papers citing a specific work, or cluster to find all versions of a paper. For US court opinions and case law, use google_scholar_cases instead.

  Connector

• verify_references

  OpenWarrant — Document Verification Suite

  Fact-check a document's REFERENCES and CLAIMS — built for AI-generated reports whose citations must be checked before they're trusted. USE THIS WHEN someone shares a report, article, whitepaper, or deep-research export (or a link to one) and asks: is this accurate / legit? are these citations real? fact-check this. did the AI make this up? Also use it proactively before relying on any AI-written document. Provide the document ONE way: `url` (a public http(s) link to a PDF or web page — fetched server-side, the cheapest call: no need to download or encode anything), `text` (pasted markdown/plain prose), OR `bytes_b64` (a base64 PDF; URLs are read from the PDF's link annotations, so they're exact). Default (fast): provenance (is it a ChatGPT deep-research export?), citation resolution (live / archived / dead, papers matched against arXiv/Crossref to catch 'real ID, wrong paper'), and internal MATH (recompute the doc's own arithmetic). Set `deep=true` to also fetch each cited source and judge whether it SUPPORTS or CONTRADICTS the claim (slower, ~a minute). Returns a trust summary, per-item tables, and a shareable `permalink` to the public fact-check record. HONEST BOUNDARY: this reports verification COVERAGE, not truth — 'supported' means evidence-backed (not necessarily true) and 'unsupported' means no evidence found (not necessarily false). It tells a reviewer WHERE to look; it does not bless the document, and it never affects the fraud risk band.

  Connector

• search_markets

  Futuur

  Search Futuur prediction markets (politics, crypto, sports, science) by text. Returns open markets by default with per-outcome implied probabilities in both play-money (OOM) and real-money (USDC) modes. Keyless.

  Connector

• export_dataset

  Trillboards DOOH Advertising

  Export observation data as a structured dataset. Supports filtering by time, geography, venue type, and observation family. Applies k-anonymity (k=5) to protect individual privacy. Queries the relevant table based on the selected dataset type, applies filters, enforces k-anonymity by suppressing groups with fewer than 5 observations, and returns structured data. WHEN TO USE: - Exporting audience data for external analysis - Building datasets for machine learning or reporting - Getting structured vehicle or commerce data for a specific time/place - Creating cross-signal datasets for correlation analysis RETURNS: - data: Array of dataset rows (schema varies by dataset type) - metadata: { row_count, k_anonymity_applied, export_id, dataset, filters_applied, time_range } - suggested_next_queries: Related exports or analyses Dataset types: - observations: Raw observation stream data (all families) - audience: Audience-specific data (face_count, demographics, attention, emotion) - vehicle: Vehicle counting and classification data - cross_signal: Pre-computed cross-signal correlation insights EXAMPLE: User: "Export audience data from retail venues last week" export_dataset({ dataset: "audience", filters: { time_range: { start: "2026-03-09", end: "2026-03-16" }, venue_type: ["retail"] }, format: "json" }) User: "Get vehicle data near geohash 9q8yy" export_dataset({ dataset: "vehicle", filters: { time_range: { start: "2026-03-15", end: "2026-03-16" }, geo: "9q8yy" } })

  Connector

• verify_references

  OpenWarrant — Reference Verification

  Fact-check a document's REFERENCES and CLAIMS — built for AI-generated reports whose citations must be checked before they're trusted. USE THIS WHEN someone shares a report, article, whitepaper, or deep-research export (or a link to one) and asks: is this accurate / legit? are these citations real? fact-check this. did the AI make this up? Also use it proactively before relying on any AI-written document. Provide the document ONE way: `url` (a public http(s) link to a PDF or web page — fetched server-side, the cheapest call: no need to download or encode anything), `text` (pasted markdown/plain prose), OR `bytes_b64` (a base64 PDF; URLs are read from the PDF's link annotations, so they're exact). Default (fast): provenance (is it a ChatGPT deep-research export?), citation resolution (live / archived / dead, papers matched against arXiv/Crossref to catch 'real ID, wrong paper'), and internal MATH (recompute the doc's own arithmetic). Set `deep=true` to also fetch each cited source and judge whether it SUPPORTS or CONTRADICTS the claim (slower, ~a minute). Returns a trust summary, per-item tables, and a shareable `permalink` to the public fact-check record. HONEST BOUNDARY: this reports verification COVERAGE, not truth — 'supported' means evidence-backed (not necessarily true) and 'unsupported' means no evidence found (not necessarily false). It tells a reviewer WHERE to look; it does not bless the document, and it never affects the fraud risk band.

  Connector

• sample_report_fetch

  refpro-mcp

  Fetch a sanitized public sample section from Refpro's reference deal library. Inputs: deal_type (FF | BRRRR | NC) and section (summary | financials | risk_notes | full). Returns sanitized example markdown content for the requested section, plus a deep-link URL to the canonical version on refpro.ai. The 'full' section stitches summary, financials, and risk_notes in order. All content is sanitized example data — not a real customer deal — and is safe to surface verbatim to end users. No network calls; samples are loaded once at module init.

  Connector

• searchCollaborators

  qubitsok — Quantum Computing Jobs, Papers & Researchers

  Find quantum computing researchers and potential collaborators from 1000+ active profiles. Use when the user asks about specific researchers, who works on a topic, or wants to find collaborators. NOT for jobs (use searchJobs) or papers (use searchPapers). AI-powered: decomposes natural language into structured filters (tag, author, affiliation, domain, focus). Returns profiles with affiliations, domains, publication count, top tags, and recent papers. Data from arXiv papers published in the last 12 months. Max 50 results. Examples: "quantum error correction researchers at Google", "trapped ions", "John Preskill".

  Connector

• searchPapers

  qubitsok — Quantum Computing Jobs, Papers & Researchers

  Search quantum computing research papers from arXiv. Use when the user asks about recent research, specific papers, or academic topics in quantum computing. NOT for jobs (use searchJobs) or researcher profiles (use searchCollaborators). Supports natural language queries decomposed via AI into structured filters (topic, tag, author, affiliation, domain). Date range defaults to last 7 days; max lookback 12 months. Returns newest first, max 50 results. Use getPaperDetails for full abstract and analysis of a specific paper. Examples: "trapped ion papers from Google", "QEC review papers this month", "quantum error correction".

  Connector

• record_agent_usage

  Push Realm

  Record that an existing learning solved your task (anonymous usage signal). Use when: • You found a learning in search results • It helped solve your problem • The solution worked as described This increments agent_usage_count by 1, which drives ranking and surfaces high-signal solutions for future agents. Call immediately after applying a solution that worked.

  Connector

• get_testing_guidelines

  IA-QA — 130+ QA & Dev Tools for AI Agents

  Query the IA-QA methodology knowledge base. Returns structured testing guidelines, assertion strategies, thresholds, best practices, and relevant MCP tools for a given topic. Call without a topic to list all available topics. Topics: llm-unit-testing, rag-pipeline, prompt-stability, prompt-ab-testing, embedding-quality, eval-framework, semantic-testing, auto-testing, security, api-testing, ci-cd, multimodal, llm-data-security, agent-observability, pro-tips, learning-paths, golden-dataset.

  Connector