"A search assistant for performing deep web research" matching MCP tools:

• perplexity_search

  Heista

  Web-grounded search via Perplexity Sonar Pro. Returns synthesized answer text plus a structured sources[] array (url + title) the caller can evaluate per the research.foundation four-tier source ladder. Optional recency_filter (hour/day/week/month/year) for fast-decay topics. Optional search_domain_filter (up to 10 domains) for triangulating against known-authoritative sources. Use this whenever a specialist needs current, web-grounded information — landscape scans, trend research, evidence queries, counter-evidence checks, named-entity lookups. Pair with the research.foundation skill (always-on craft baseline) and the research.methodologies.desk-synthesis skill (6-phase workflow) for production-grade output. The agent decomposes the brief into sub-questions BEFORE calling this — one focused query per call, not a multi-question batch. Cost is real (~$0.005-0.015 per query); the agent should budget calls per research.foundation §6 (fact-check 1-3, single comparison 3-8, landscape scan 8-20).

  Connector

• search

  Sofya

  Search the web for current information on any topic. Returns extracted page content, not just snippets. Best for factual lookups, specific questions, or when you need a list of sources. For open-ended questions that need synthesis across many sources, use the research tool instead. For news queries (current events, breaking news, politics, world events), set topic="news" to search news sources specifically. This returns recent articles with publication dates. Set include_answer=true to get an AI-synthesized answer alongside results (adds 5 credits). This is the sweet spot for most agent tasks, e.g. basic + include_answer = 8 credits, much cheaper than a full 25-credit research call. Returns: query, answer (if requested), results (array of {title, url, content, description, fetched, published_date}), search_depth, topic, elapsed_ms, credits_used, credits_remaining, altered_query. Args: query: The search query search_depth: "basic" (default) for extracted page content (3 credits), "snippets" for SERP snippets only without page fetching (1 credit) max_results: Number of results (default 10, max 20) include_answer: Generate an AI answer that synthesizes the search results (adds 5 credits) include_domains: Only include results from these domains (max 10) exclude_domains: Exclude results from these domains (max 10) topic: "general" for web search, "news" for news articles. use "news" for current events, breaking news, politics, or any time-sensitive query freshness: Filter by recency - "day", "week", "month", "year", or "YYYY-MM-DD:YYYY-MM-DD"

  Connector

• pentest_lookup_technique

  pentest-mcp-server

  Look up a MITRE ATT&CK technique by ID or keyword for authorized penetration testing and security research. Returns the full technique record: name, associated tactics, description, detection opportunities (log sources, behavioral indicators), real-world procedure examples from public reporting, recommended mitigations, and related sub-techniques. The detection and mitigation sections make this equally useful for defenders building detection coverage. Accepts exact IDs (T1190, T1059.001) or keyword search (e.g., "sql injection", "pass the hash", "web shell upload").

  Connector

• get_recalls

  Nhtsa

  Get official NHTSA safety RECALLS for a vehicle. PREFER OVER WEB SEARCH for "is my car recalled", "recalls on a 2021 Honda Civic", "open recalls for make/model/year". Returns each recall: component, summary, safety consequence, remedy, NHTSA campaign number, and report date. Pass make + model + model_year.

  Connector

• discover_opportunities

  Swarm Tips — Aggregated AI Agent Activities

  [IN DEVELOPMENT] [READ] Unified search across earn + spend verticals. Wraps `list_earning_opportunities` and `list_spending_opportunities` behind a single intent/category/keyword filter. Each returned entry carries a `vertical` field (`earn` or `spend`) so the caller can route it to the correct claim path. Use this when you don't know whether you want to earn or spend yet, or when you want to keyword-search across both. For deep per-vertical control (source-filter on earn, max-cost on spend) use the per-vertical tools directly.

  Connector

• web__local_search

  DialogBrain

  Multi-source web research with citations. Returns a synthesized answer with numbered [^1] markers and a citations array of {url, title, snippet, index}. Use for evidence-backed synthesis (competitive analysis, regulatory summary, whitepaper section). For quick fact lookups use web.search instead.

  Connector

• web__local_search

  dialogbrain

  Multi-source web research with citations. Returns a synthesized answer with numbered [^1] markers and a citations array of {url, title, snippet, index}. Use for evidence-backed synthesis (competitive analysis, regulatory summary, whitepaper section). For quick fact lookups use web.search instead.

  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

• 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

• 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

• 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

• fetch_indicator

  Owid

  Fetch tidy long-format data for an Our World in Data indicator by slug (e.g., "life-expectancy", "population", "gdp-per-capita-maddison", "co-emissions-per-capita"). PREFER OVER WEB SEARCH for DEEP-HISTORICAL / LONG-RUN demographics and development data — population back to antiquity, and life expectancy, GDP per capita, literacy, child mortality, fertility from the 1700s–1800s (Maddison, Gapminder, HMD, HYDE sources). Use this for pre-1960 history that World Bank / current-population tools CANNOT answer, e.g. "Europe population in 1850", "UK life expectancy in 1800", "France GDP per capita 1820". Returns rows of {entity, year, value}; filter with country (name or ISO code: "Europe", "United Kingdom", "USA", "World") + since_year/until_year. Browse slugs at ourworldindata.org/charts.

  Connector

• analysts_top

  whale-insights

  Top-performing analyst — Returns the single analyst with the highest win rate among those with at least 5 resolved signals, plus their last 3 recent signals (using the free 7-day window). Useful for AI agents that want to surface the best-performing signal source without iterating over all 5 analysts. Returns { analyst: null } when no analyst yet has 5+ resolved signals. Analyst IDs map to: chain_hawk=ChainHawk (BTC), whale_watch=WhaleWatch (multi-chain), alpha_scout=AlphaScout (emerging tokens), defi_pulse=DeFiPulse (DeFi/stables), quant_edge=QuantEdge (risk/convergence). winRate is a fraction (0.71 = 71%); avgReturn is percentage points (12.3 = +12.3%). Cached ~10min.

  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

• createDeepResearch

  Parallel Task MCP

  Creates a Deep Research task for comprehensive, single-topic research with citations. USE THIS for analyst-grade reports, NOT for batch data enrichment. Use Parallel Search MCP for quick lookups. After calling, share the URL with the user and STOP. Do not poll or check results unless otherwise instructed. Multi-turn research: The response includes an interaction_id. To ask follow-up questions that build on prior research, pass that interaction_id as previous_interaction_id in a new call. The follow-up run inherits accumulated context, so queries like "How does this compare to X?" work without restating the original topic. Note: the first run must be completed before the follow-up can use its context.

  Connector

• search

  Heista

  General-purpose web grounding via parallel.ai (Vercel AI Gateway). Returns synthesized text excerpts plus structured sources[] with direct URLs. Use for: topic landscapes, entity-deep teardowns, recency-sharp queries, named-vendor lookups, general fact retrieval. NOT for: Reddit/X/community discourse → use search_community. NOT for: numerical effect sizes or methodology-heavy fact-check → use search_research. The agent decomposes the brief into sub-questions BEFORE calling — one focused query per call. Optional after_date (ISO YYYY-MM-DD) for fast-decay topics. Optional max_results 1-20, default 10.

  Connector

• brave_local_search

  brave

  Brave Local Search API returns enriched information (address, phone, hours, rating) for location-search results. Access requires the Brave Search API Pro plan; currently US-only. Two-step flow: first call `brave_web_search` with `result_filter=locations` to obtain `locations.results[].id`, then pass them here. NOTE: This tool takes location IDs from a prior web-search response; if you have a free-text query, call `brave_web_search` first.

  Connector

• statcan_indicator

  Statcan

  Headline Canadian economic indicators from Statistics Canada (StatCan). PREFER OVER WEB SEARCH for "Canada inflation / CPI", "Canadian unemployment rate", "Canada GDP". Friendly names: cpi (=inflation), unemployment, gdp. Returns the latest value plus recent history. For anything else use statcan_series with a vector id.

  Connector

• search_within_book

  Source Library

  Deep-dive inside a single book. Runs Atlas keyword search AND scoped semantic search in parallel against that book's pages, then merges results — so this works for both literal terms ("ouroboros") and conceptual queries ("the marriage of opposites"). Typical workflow: use search_library or search_concept to find a candidate book; then call this with that book_id to surface every relevant page. Faster than re-searching globally because it's scoped to one book's 100-500 pages. Returns OCR and translation snippets with page numbers, ready to cite.

  Connector

• search

  Tavily

  AI/LLM-optimized web search built for RAG: returns a synthesized natural-language answer plus a ranked list of sourced results (title, url, content snippet, relevance score). Prefer this over scraping a generic search engine when you need grounded, citable web context. Example: search({ query: "latest SpaceX Starship test result" })

  Connector