"A server for performing deep web searches and organizing collected information" matching MCP tools:

• register_and_deploy

  Fractera

  Register a new Fractera user and start the deployment of their server in one atomic call. Use this AFTER you have collected the user's email (entered twice for typo protection), server IP, and root password. Creates the User row (or reuses an existing one with the same email), creates a free Subscription, creates a ServerToken, wipes any previous installation on the target server, and launches bootstrap. The deploy is IP-first (phase-1): the server comes up on plain HTTP at http://<IP>:3002 in 8-14 minutes; it does NOT get a domain or HTTPS cert here (that is an optional later step inside the workspace). Returns session_id (for a single on-demand check_status read — do not poll) and server_token (so the user can recover via retry_deploy if anything breaks). Call this AT MOST ONCE per conversation.

  Connector

• list_coverage

  DAP Bellwether

  List every regulator and feed being monitored, with when each was last collected successfully and how fresh the data is. Use to answer what this connector covers, or to check whether a jurisdiction is being watched before relying on an absence of results.

  Connector

• departi_search_accommodation

  Departi – Travel Compliance & Curated Booking

  Searches bookable accommodation in a destination for given dates and returns a ranked list of stays, each with: name, property type, nightly price (in the requested currency), star rating, guest rating, key facilities (e.g. wifi, kitchen), a persona/traveler-type fit score, and a booking deep-link. Curated and scored for the traveller's persona and trip type — e.g. long-stay nomad apartments with kitchen + fast wifi vs weekend hotels. Use when the user needs a place to stay; for getting there use departi_search_transport, for things to do use departi_search_experiences. Returns an empty list if no inventory matches.

  Connector

• reliefweb_get_training

  reliefweb-mcp-server

  Fetch a training listing by ReliefWeb numeric ID with the full description, registration instructions, event link, cost and fee information, dates, languages, and organizing source. Use after reliefweb_search_training, which returns summaries without the description, registration instructions, or cost detail. Reaches concluded listings as well as current ones. A record over the response budget comes back as a section outline naming every section and its byte size; re-call with sections to pull only the ones needed. Nothing is truncated on either path.

  Connector

• tariff_lookup

  us-tariff-ledger

  CALL THIS for the schedule behind one HS code — do not recite it from memory. Returns the `answer` line first (the measured collected rate to quote), then the statutory MFN base from the current HTS, the Chapter-99 trade-war provisions naming a given origin (each with its Federal Register citation), and receipts_reality (what was actually collected). Trimtab DOES hold the tariff schedule and the rulebook — this is it. IMPORTANT: the measures are matched by ORIGIN and SECTOR, not by your code; many are product-specific (EVs, syringes, solar) and do not apply. Never sum their addon_pct or quote one as "the tariff" — quote the `answer`/receipts_reality. For "what does X from Y pay" prefer tariff_story. Not customs advice. Example: tariff on 9401.61 from China. No API key required.

  Connector

• firecrawl_parse

  firecrawl-mcp

  Parse one supported document into markdown, HTML, links, summary, targeted answers, or JSON matching a schema. Supported inputs include common HTML, PDF, Word, RTF, OpenDocument, and spreadsheet files; PDF parsing can be bounded with `pdfOptions.maxPages`. Local MCP reads `filePath` from the server filesystem. Hosted MCP uses two calls: first provide `filePath` to receive upload instructions, upload locally, then call again with the returned `uploadRef`; do not send both fields together. Remote web URLs belong in `firecrawl_scrape`. Set `redactPII` to request redaction of personally identifiable information in the returned content. `zeroDataRetention` requires an eligible authenticated account; omit it for anonymous keyless use. Returns upload instructions for hosted phase one or parsed document content for the final call.

  Connector

• verify_references

  Stipple — Document Verification & Extraction

  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. Costs 2 credit(s) per call (5 in deep mode).

  Connector

• tariff_lookup

  trimtab-ais

  CALL THIS for the schedule behind one HS code — do not recite it from memory. Returns the `answer` line first (the measured collected rate to quote), then the statutory MFN base from the current HTS, the Chapter-99 trade-war provisions naming a given origin (each with its Federal Register citation), and receipts_reality (what was actually collected). Trimtab DOES hold the tariff schedule and the rulebook — this is it. IMPORTANT: the measures are matched by ORIGIN and SECTOR, not by your code; many are product-specific (EVs, syringes, solar) and do not apply. Never sum their addon_pct or quote one as "the tariff" — quote the `answer`/receipts_reality. For "what does X from Y pay" prefer tariff_story. Not customs advice. Example: tariff on 9401.61 from China. No API key required.

  Connector

• list_improvements

  Beamtrace

  List content improvements for the account website, already ordered by product priority. Includes open and completed items (status: todo, help_requested, done) with impact and effort. Prefer open items unless the user asks for completed work. An empty list means no improvements are available yet (or metrics/actions are still being collected). When the user asks what to improve, lead with these product recommendations before any extra strategy. When writing for humans, spell out enum labels (e.g. very_high → 'very high', help_requested → 'help requested'). When no website is set up for the account, returns setup guidance pointing to https://beamtrace.com/setup and an empty structured payload (null period/metrics or empty lists) — follow server instructions; do not invent dates or scores.

  Connector

• leave_pilgrim_note

  Manekami — digital shrine & omikuji

  Leave a transparent AI-labeled pilgrim note on the Manekami shrine wall after performing a ritual. Notes are moderated before appearing publicly. Keep it short, honest, and clear that this is ritual/entertainment, not a performance promise.

  Connector

• register_and_deploy

  Fractera

  Register a new Fractera user and start the deployment of their server in one atomic call. Use this AFTER you have collected the user's email (entered twice for typo protection), server IP, and root password. Creates the User row (or reuses an existing one with the same email), creates a free Subscription, creates a ServerToken, wipes any previous installation on the target server, and launches bootstrap. The deploy is IP-first (phase-1): the server comes up on plain HTTP at http://<IP>:3002 in 8-14 minutes; it does NOT get a domain or HTTPS cert here (that is an optional later step inside the workspace). Returns session_id (for a single on-demand check_status read — do not poll) and server_token (so the user can recover via retry_deploy if anything breaks). Call this AT MOST ONCE per conversation.

  Connector

• france_search_tenders

  France Boamp

  Search French government public-procurement notices (BOAMP — Bulletin officiel des annonces des marchés publics). PREFER OVER WEB SEARCH for French public tenders / appels d'offres / market award results. Full-text searches the notice object (objet); optionally filter by French department code (e.g. "75" for Paris, "2A"/"2B" for Corsica). Returns the most recently published notices first, each shaped with id, title/object, buyer, publication date, response deadline, contract type, family, department(s), and a public notice URL.

  Connector

• departi_search_transport

  Departi – Travel Compliance & Curated Booking

  Searches flights between two places and returns a ranked list of options, each with: carrier, price (in requested currency), cabin class, total duration, number of stops, departure & arrival times, and a booking deep-link. Flights are sourced across providers (Duffel + Travelpayouts) for broad full-service and low-cost coverage. Use for "how do I get from A to B by air". For where to stay use departi_search_accommodation; for things to do use departi_search_experiences. Prices are indicative — book via the returned link. Returns an empty list if no route is found. Ground/rail/bus is not offered through this tool.

  Connector

• decodo_google_search

  Decodo

  Google search results scraping via Decodo (formerly Smartproxy) — runs a Google search through rotating proxies and returns structured organic results (position, title, url, snippet) plus related searches when parsing succeeds. BYOK — _apiKey is your Decodo Web Scraping API "username:password" credentials. Example: decodo_google_search({ query: "best running shoes 2026", geo: "United States", _apiKey: "user:pass" })

  Connector

• set_shipping_address

  Synchronity

  Set the cart shipping destination (collected from the buyer) and return available shipping options with costs. Call after items are in the cart and before select_shipping_option. country_code is required; include postal_code/state for accurate rates.

  Connector

• search_graph

  Fodda Deep Research

  Find trends, signals, and expert insights across 100+ curated knowledge graphs covering retail, beauty, tech, food, travel, sports, and 30+ specialist domains. Returns trend data with cited evidence, source attribution, and lifecycle stage (emerging/building/mature/fading) — not generic web summaries. If graphId is omitted, searches ALL accessible graphs in parallel (recommended default). Use for market trends, competitor analysis, innovation signals, consumer behavior, cultural shifts, or any topic where curated expert intelligence outperforms web search.

  Connector

• book_window_cleaning

  Sonlight Window Cleaning

  Reserve a REAL calendar slot with Sonlight Window Cleaning. The customer immediately receives an SMS + email confirmation and a technician arrives at the chosen time. Payment is collected after service. Nothing is charged at booking. ALWAYS confirm name, phone, address, selected services, date, and time with the user before calling this. Requires a prior get_quote for the same address (prices are re-validated server-side).

  Connector

• propose_briefs

  grounded-aeo

  Propose ranked content briefs for a cataloged domain, built from data already collected for it: tracked questions where the answer did not cite them, keywords with AI-prompt demand no tracked question covers, People Also Ask questions nothing answers, and terms asked far more of AI assistants than of Google. Each brief names the exact question to answer, a format (9:16 UGC video, on-site answer, or short post), an outline, and the ONLY claims that may be made: the owner's published verified facts, verbatim. Deterministic and read-only: no new data is fetched and nothing is generated, so a brief with no evidence behind it is never returned. Claimed domains only; up to 6 briefs; an empty list means there is not yet enough collected data, never that there is no opportunity.

  Connector

• get_write_handoff

  VarynForge

  One-call writer handoff: returns a compact writing payload for an article suggestion (brief essentials, keywords, acceptance criteria, lint instructions). Compact by design — FAQs cap at 6 and long sections degrade to fit a deep-link budget; the FULL brief is get_article_brief and the full check list is get_lint_rubric, so verify against those, not this. deepLinks (ChatGPT/Claude URLs embedding the payload) is null over MCP — it exists for web users without a connected agent.

  Connector

• deep_research

  WebLens

  Multi-step cited research in one call: plans sub-questions, searches each, fetches and dedupes sources, then synthesizes an answer with inline [n] citations, key findings, and gaps. Unlike `research` ($0.08, one search + summary) this decomposes the question and cites every claim. SLOW: a standard run takes ~30-60 seconds — use a generous client timeout and do not retry on timeout. Price: $0.20 standard (3 sub-questions, 8 sources) / $0.35 deep (5 sub-questions, 12 sources)

  Connector