ConnectorNo auth
726,822 tools. Updated 2026-10-30 23:38
"replicate" matching MCP tools:
- Find arbitrage opportunities on Polymarket via monotonicity violations + partition-sum checks. Call with NO args for a `trending_scan` of the top ~200 markets by weekly volume; pass `event` for the strongest per-event partition_check, or `topic` for a themed cross-event scan. `event` (recommended for a specific market): pass a Polymarket event slug like "fed-decision-may-2026" or "when-will-bitcoin-hit-150k"; walks child markets, checks date-axis / threshold-axis ordering AND computes the partition_check (sum of YES prices across mutually-exclusive legs — should ≈1; deviations >3pp emit a BUY/SELL EVERY LEG signal). `topic` (for cross-event scanning): pass a seed question like "Strait of Hormuz traffic returns to normal" or "Fed rate decision"; searches related events across the platform, flattens markets, runs the comparator on the union. Cross-event mode catches "...by May 31" vs "...by Jun 30" patterns that single-event misses. SEMANTIC ANCHOR: cross-event pairs require ≥0.30 Jaccard similarity on question tokens (prevents Powell-Fed-Pause being paired with Powell-DOJ-probe); skipped_low_similarity surfaces the rejected pair count. PARTITION FILTER: drops will-person-X / will-manager-Y / will-someone-else- placeholder slugs; partitions with >20% placeholder fraction return null arb signal. Response: opportunities[] (gap_pp, suggested_trade, reasoning, monotonicity violation context), and in event mode partition_check{sum_yes_prices, gap_from_1, placeholders_filtered, suggested_trade}. FEES: every opportunities[] row and partition_check.arbitrage carry edge_pp_gross (== gap_pp / overround_pp), fees_pp, edge_pp_net, net_positive, plus polymarket_fee_pp, fee_basis and fee_categories[]. BOTH cost components are modeled: Polymarket's own per-category TAKER FEE (fee = shares × rate × p × (1-p), rates crypto 0.07 / sports-economics-culture-weather-other 0.05 / finance-politics-mentions-tech 0.04, geopolitics and world events fee-free; verified against Polymarket's own docs as of 2026-09-13) and Polygon gas (~$0.02/leg). The taker fee dominates: ~$1.75 per 100 shares on a crypto market at 50c versus $0.02 of gas, so rows that looked profitable before fleet #1927 may now show net_positive:false — that is the correction, not a regression. Each leg is priced at ITS OWN market's rate and price (the fee curve peaks at 50c and falls toward both extremes). fee_basis says where the rate came from: 'payload' (read off the market, the normal case), 'category' (mapped from its fee category), 'fee_free', or 'fallback' (rate unknown — charged at the modal 0.05 rather than assumed free, so an unreadable market is never reported as costless). Where fill_check reprices against live depth, this does NOT double-count that spread cost. FILL CHECK: when the partition signal fires, arbitrage.fill_check prices it against live CLOB depth (theoretical_edge_pp_at_book vs realizable_edge_pp at 1000 shares/leg, thin_legs[]) — realizable_edge_pp ≤ 0 means the overround exists only at last-trade, not in the book; do not trade it. For custom sizing use polymarket_fill_risk.ConnectorNo auth
- Read ONE ad in full, live from Meta — its creative content (primary texts, headlines, descriptions, CTA, destination link, URL parameters, image hashes / video ids, Facebook Page and Instagram identity) plus status, timestamps, object_story_id / effective_object_story_id (the underlying post, which carries the ad's likes and comments) and the raw creative specs. Get the ad_id from adsap_list_ads or adsap_get_insights. Use it to inspect an ad's copy and media, to replicate an ad, or to answer 'what does this ad say / link to?'. Read-only — to CHANGE a live ad's creative use adsap_replace_ad_creative.ConnectorAPI key
- Copy an Elementor page EXACTLY from ONE site to ANOTHER, server-side. Reads the source's raw builder JSON inside Xyle (never through your context, so nothing is truncated) and writes it verbatim to a new page on the destination — every section, text string and style preserved, nothing dropped. Use this for 'replicate this page onto site B' instead of reading with wp_get_page and rebuilding (a large page's data is hundreds of KB and cannot be reproduced losslessly by hand). Optionally re-hosts the source's images into the destination's media library so the copy is self-contained. Elementor Pro widgets are copied as-is and reported (they need Pro on the destination to render). Creates a DRAFT by default; the source is untouched. For a copy WITHIN one site use wp_clone_page instead.ConnectorOAuth
- Prices Kalshi daily high-temperature markets against the NWS forecast for the market's OWN settlement station, and measures whether that forecast actually beats the market. Two modes. LIVE (default): returns the full strike ladder for one city and settlement date with market_prob (mid), forecast_prob, and edge_pp per strike, plus the settlement clause verbatim. BACKTEST (`backtest_days: N`): scores an archived gridded forecast against the market on settled days and returns brier_market vs brier_forecast with a plain-English `verdict`, so the edge is MEASURED rather than asserted. READ THE WARNINGS — they are not boilerplate. (1) These markets DO NOT settle on the NWS. They settle on The Weather Company (weather.com) at a Kalshi station code such as CLINYC, which the response quotes verbatim; so part of every edge_pp is NWS-vs-Weather-Company disagreement about the same day at the same station, which is not mispricing and not tradeable. `settlement_vs_forecast_basis_f` from backtest mode is that part as a number. (2) The station is DERIVED from the settlement clause, never from the city name: Chicago settles at MIDWAY and New York at CENTRAL PARK, so a city-centre forecast would misprice a whole ladder. A station that cannot be resolved yields rows with no forecast and a reason, never a guessed coordinate. (3) forecast_prob assumes a normal distribution around the NWS high whose width is ASSUMED, not fitted (stated in `distribution_assumption`) — run backtest mode to see whether it is calibrated. (4) edge_pp is gross: no Kalshi fees, no bid-ask. MEASURED RESULT, AND IT IS NOT THE FLATTERING ONE: on the first backtest (KXHIGHNY, 13 settled days to 2026-09-11, 58 market observations) the MARKET beat the forecast — Brier 0.1008 for the market against 0.1594 for the archived gridded forecast, lower being better. So on that sample there is NO forecast edge to sell, and a large edge_pp is more likely to be the model disagreeing with a better-informed market than an opportunity. The measured settlement-vs-forecast basis was 1.7F mean absolute over 8 pinnable days, slightly warm-biased, which is a big share of a typical edge_pp on a 2-degree bracket. Re-run backtest_days before believing any edge; if a later sample reverses this, the numbers say so. NWS is US-only, so the ~30 international Kalshi weather series (London, Paris, Tokyo) return market prices with forecast_unavailable rather than a forecast. Precipitation series are listed but not yet priced. Cities: nyc, chicago, los angeles, miami, austin, houston, denver, philadelphia — or pass `series_ticker` for any other (e.g. "KXHIGHTBOS").ConnectorNo auth
- Claygent-style per-row research: answer ONE question about ONE company and return a JSON object matching your output_schema, with every field wrapped in its own evidence — {value, evidence_url, quote, status}. status is NEVER self-reported by the model: "found" means this code independently verified the model's quote is a VERBATIM substring of text actually fetched; "inferred" means a value was given but the quote could not be verified; "not_established" means no source answered the field, and its value is always null — never guessed, never a blended score. Sources: "web" fetches the company's own site (entity.domain required — homepage, /about, /leadership) via fetch_url; "catalog" checks entity_profile (SEC EDGAR; resolved:false for a private company is expected, not a failure); "both" (default) tries both. max_cost (USD, default 0.02, max 0.50) scales how many pages are fetched and refuses up front if it is too small to pay for even one extraction call (floor $0.002) — never silently overspends. A row where every source comes back empty costs nothing: every field is mechanically not_established with NO model call made. For a boolean/derived field (e.g. "does this company sell to healthcare"), the quote only has to SUPPORT the value — a descriptive sentence about customers/products counts, the source need not state the answer verbatim. leadership_page_status ("fetched"|"not_found"|"not_attempted", plus leadership_note) says plainly whether a leadership/team page was actually read, so a named-person field (head_of_sales) coming back not_established is distinguishable from "this company just doesn't publish one" vs. "we never checked". model_output_empty:true means the model returned nothing usable (a transport/budget problem) — every field is not_established for THAT reason, never read as 20 individual "no" judgments. Use with a fixed question + schema across many rows for "Claygent over N companies with a hard cost cap" (composes with batch callers). For picking from a fixed label list instead of extracting open fields, use classify_row (cheaper, one model call).ConnectorNo auth
Matching MCP Servers
- AlicenseAqualityCmaintenanceAn MCP server that enables AI assistants to search, discover, and run any model on Replicate.5MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to run open models such as Flux, Llama, and Whisper and to manage the resulting predictions, deployments, and model metadata. Supports starting, polling, streaming, and cancelling runs, with webhook notifications for completion.2,071 npmMIT
Matching MCP Connectors
Replicate MCP.
Run AI models, create deployments, and manage predictions via cloud API
- Tell the Pipeworx team something is broken, missing, or needs to exist. Use when a tool returns wrong/stale data (bug), when a tool you wish existed isn't in the catalog (feature/data_gap), or when something worked surprisingly well (praise). ONLY for tools served by this Pipeworx connection — if the tool came from a different MCP server in your client (another vendor's Gmail, Splunk, Slack, etc. connector), we cannot fix it and reporting it here only delays you; file it with that server instead. Not sure? Pipeworx tool names are the ones this connection lists. Describe the issue in terms of Pipeworx tools/packs — don't paste the end-user's prompt. Filing without an account returns a `claim_token`; pass it back later as pipeworx_feedback({claim_token:"pwfb_…"}) to read whether it was fixed and what changed. The team reads digests daily and signal directly affects roadmap. Rate-limited to 5 per identifier per day. Free; doesn't count against your tool-call quota.ConnectorNo auth
- IMPORTANT — bulk domain migration: domain moves via revise are for individual corrections only. If the user needs to move many memories between domains, inform the user that bulk migration must be performed via the admin interface (merge_domains) — do not attempt to replicate a merge by looping revise calls. Update one or more existing memories. Omitted fields are unchanged. Single: pass fields directly — returns {updated, connections, suggested_connections, applicable_standing_rules} and, when the filing-time threshold is crossed, possible_contradicts + possible_contradicts_candidates (same shapes as remember()). Batch: {items:[{id,...},...]} — returns {items:[{id, updated, connections, suggested_connections, applicable_standing_rules, ...}]} with a per-item envelope on each success, not only an updated count. After every successful revise — plain update, override, claim, or supersede — review applicable_standing_rules (domain-scoped standing rules, distance-independent), connections, suggested_connections, and possible_contradicts in the same turn; do not defer to a separate recall or suggest_connections call. Semantic similarity reflects aboutness, not agreement; the server surfaces candidates that may warrant your review but never asserts they conflict. Every memory has an owner (whoever created it) — revising a memory you don't own is rejected with error_class=forbidden unless you are an Editor or Owner and supply override_reason (required) plus override_confirm=true (required only when the memory is human-owned; not required for agent-owned or ownerless memories). The override path requires a session with a workspace role (a human JWT session, or a personal key linked to an Editor/Owner user); sessions on plain workspace keys cannot override regardless of arguments. An override changes content in-place, not ownership — substantive changes (label, description, why_matters, node_kind) on a foreign-owned memory are rejected; use supersede=true instead, which creates your successor memory, archives the original intact, and wires a supersedes relationship. supersede=true (single-memory form only) returns {superseded, archived_id, archived_connections} plus the revise envelope on the successor; foreign supersede uses the same override_reason/override_confirm ceremony. Correction-class overrides (tags, occurred_at, transient) may still use in-place override. override_reason/override_confirm/supersede apply to the single-memory form only — batch revise has no override or supersede path: if any item in the batch targets a memory you don't own, the whole batch is rejected and none of it applies; revise that item individually instead. claim=true (single-memory form only) makes an ownerless memory (owner_id IS NULL — either it predates ownership tracking, or was orphaned by a member offboard) yours: requires Editor or Owner role, is rejected with error_class=validation if the memory already has an owner (use override or supersede instead), and error_class=conflict if someone else claimed it first (race). claim never moves a memory from one owner to another — only from no owner to you — and may be combined with other field updates in the same call. domain (single-memory form only) moves the memory to a different domain; domain_move_reason is required when domain is present — the call is rejected with error_class=validation if domain_move_reason is absent; domain equal to the memory's current domain is also rejected. On failure, content[0].text is JSON: {"error_class": "not_found|conflict|retryable|forbidden|validation|internal", "message": "..."}. Switch on error_class: retry on retryable, surface message on validation, re-fetch on not_found.ConnectorAPI key
- List issues the AI found while analyzing session recordings: bugs, broken flows, and UX problems, deduplicated across sessions and ranked by severity (or by last seen with sort recency). Each row has title, severity, status, sessions affected, and first/last seen; the response also carries site-wide open, in_progress, and resolved counts plus pagination.total. Start here for 'what is broken', then call get_issue with an id for occurrences, steps to replicate, and comments. Suspended issues are hidden unless status is suspended, so an issue that vanished was probably suspended, not deleted. On a free trial only the first 10 issues are listed and counted. Limit defaults to 100 (max 1000). Requires websiteId or domain with a workspace token.ConnectorOAuth
- Get full detail for one AI-detected issue: every occurrence with timestamps, the sessions behind it, steps to replicate, comments, and any linked Linear or Jira ticket. Get issueId from list_issues; use this only when you need the evidence behind one issue. An unknown id, or one from another website, fails with 'Issue not found'; on a free trial only the first 10 issues are available, and later ones fail as not included in the current plan. Each session lists its recording id, the anonymous visitorUid, start URL and page title, timing and event counts; no names, emails or locations are returned, so surface only what answers the question. Requires websiteId or domain with a workspace token.ConnectorOAuth
- Render a batch of TikTok photo carousels from a template: swipeable 1080x1920 slides, real library photos, the copy composited as REAL text (never generated into the image, so it says exactly what you write). ASYNC — returns a jobId; poll get_job until completed, and the slide URLs come back on the job itself. SPENDS 1 CREDIT PER CAROUSEL. Max 12 per call. Keys in `values` must come from list_carousel_templates; unknown ones are reported in ignoredKeys, not rejected. Image slots take either a `collection:<pool>` key (stock photos, see the template's examples) or a direct image URL. For candid phone-shot backgrounds that match each slide's line, generate them first with studio_generate_image (modelKey 'replicate-pruna-p-image-ideogram', 1 credit, aspectRatio '9:16', prompt like 'candid iPhone photo of a laptop showing a blurred inbox at night, phone-camera grain, no people, no readable text, no logos') and pass the returned URL.ConnectorDestructiveOAuth
- Duplicate an element and its whole subtree N times as siblings. THE way to build a row of cards, a pricing grid, a team list, a logo strip. Build ONE to spec, screenshot it, then replicate and swap only the content. Hand-authoring six cards produces six slightly different cards — each is another chance to typo a hex or miss a radius, and nothing holds them together afterwards. A verified source copied six times cannot drift. Every id is re-issued (elements AND repeater item _ids), because Elementor keys per-element CSS off them: duplicated ids leak styling between copies and make editing one silently change another. Response returns id_maps (original id -> new id per copy) so you can patch a specific copy's insides without re-reading the page.ConnectorDestructiveOAuth
- "Compare X and Y" / "X vs Y" / "X versus Y" / "which is bigger / better / larger / more profitable" / "rank these companies" / "head to head" — side-by-side comparison of 2–5 companies or drugs in ONE parallel call. ALWAYS PREFER over sequential single-pack lookups when comparing entities. type="company" pulls LATEST 10-K revenue + operating income + net income + cash + long-term debt from SEC EDGAR/XBRL (off-calendar fiscal years handled correctly — AAPL Sep, NVDA Jan, etc.). type="drug" pulls FAERS adverse-event counts, FDA approval counts, active trial counts. Results sorted by primary metric so "largest" / "most" / "biggest" reads off the top of the response. Returns paired data + pipeworx:// citation URIs per entity. Replaces 8–15 sequential lookups.ConnectorNo auth
- "Is it true that…" / "fact check" / "verify the claim that…" / "did X really…" / "was Y actually…" / "confirm or refute" / "true or false" — natural-language claim verification against authoritative sources. Use whenever the agent needs to check whether something a user said is factually correct. Company-financial claims (revenue, net income, cash for public US companies) verify via the structured SEC EDGAR + XBRL fast path with exact percent-delta math; ANY OTHER factual claim (macro statistics, rates, prices, drug data, records) automatically falls through to the grounded pipeline — routed to the right live source, answered with verbatim evidence, then judged. Returns a verdict (confirmed / approximately_correct / refuted / inconclusive / unsupported / could_not_verify), the grounded or structured actual value with pipeworx:// citation, and reasoning. IMPORTANT for callers: could_not_verify means the check did not happen (our LLM or source failed) and carries verification_error{stage,detail} — it is NOT evidence for or against the claim, and must not be shown as one. unsupported means we looked and cover no source for it. Replaces 4–6 sequential calls (NL parsing → entity resolution → data lookup → comparison).ConnectorNo auth
- TYPED, DETERMINISTIC financial facts for a US public company for an EXPLICITLY NAMED reporting period — "Apple revenue for fiscal 2023", "Walmart net income FY2026 Q3", "Microsoft cash at the end of fiscal 2024". PREFER OVER entity_profile / get_company_financials whenever the period matters: those answer "the most recent figures" and will happily hand back FY2025 when you asked about FY2019, and neither separates a discrete quarter from a year-to-date figure. This one refuses instead — it NEVER substitutes the latest period for the period requested, NEVER returns 0 for missing data, NEVER lets a 9-month YTD number answer a quarterly question, and NEVER converts a currency. Every answer carries the exact us-gaap concept it came from, what that concept MEASURES (NetIncomeLoss excludes non-controlling interests, ProfitLoss includes them — not synonyms), the accession number and a link to the filing on sec.gov, the restatement trail of any superseded figures, and a contract + derivation version to pin against. Fiscal periods are the FILER'S OWN, anchored on their fiscal-year end, so Walmart's year ending 2026-01-31 is FY2026 and Apple's ending 2025-09-27 is FY2025. Attributes: revenue, net_income, cash, gross_profit, operating_income, eps_diluted, capex, operating_cash_flow, long_term_debt, shares_outstanding (the dei COVER-PAGE share count, so its period end is the cover date, after the quarter closed — that is the most current count the filing carries). Units are not all USD: shares_outstanding is in `shares` with currency null, eps_diluted in `USD/shares`. A discrete quarter the filer never filed — every Q4, and Q2/Q3 of cash-flow lines, which are filed year-to-date — is DERIVED as YTD(q) minus YTD(q-1) and comes back with `reported: false`, the subtraction named in `derivation` (fy_minus_9m / ytd_9m_minus_6m / ytd_6m_minus_3m) and both legs with their filings in `derived_from`; test `reported` if you will only accept a figure lifted from a filing. Per-share and balance-sheet figures are NEVER derived. Every non-answer is a named status — `unavailable` (the filer did not report it for that period; the periods that DO exist are listed, without values), `unsupported` (outside what v1 covers — a non-us-gaap filer, an unknown attribute, a non-USD unit), `ambiguous` (the company name matched two filers equally well; both are named), `conflicting` (two filings the same day disagree; both are returned and neither is picked), `partial` (a value with no accession behind it). Source: SEC EDGAR, XBRL financial facts, one publisher read once — see `corroboration`. Same response is served at POST https://gateway.pipeworx.io/v1/facts for non-MCP callers.ConnectorNo auth
- Search the NATIONAL BUSINESS REGISTERS by company name and get the register's own id back — a Companies House number, a Norwegian organisasjonsnummer, a Finnish business id, a SIREN, a Japanese corporate number, an Estonian registry code, a CNPJ, or an LEI. Use this when you want the OFFICIAL REGISTERED ENTITY and the country is not the US, or when you do not know which country it is. Registers searched: Companies House (GB), Enhetsregisteret (NO), PRH/YTJ (FI), Sirene (FR), 法人番号公表サイト (JP), RIK e-Business Register (EE), and the GLEIF LEI crosswalk, which reaches jurisdictions we have no register pack for (DE, SE, AU, SG...). THIS TOOL REFUSES TO GUESS, and that is its point: when two different companies match a name equally well it answers `match_status: "ambiguous"` with BOTH, rather than naming one. A one-word trading name usually IS ambiguous — "Bolt", "Apple", "Next", "Sea", "DNB" each match a well-known group AND unrelated micro-entities whose registered name is exactly that word — so pass `country` when you know it. Every response names EVERY register it asked under `searched[]`, with that register's own as-of date and row count, and a register that could not be reached carries an `error` there rather than reading as zero rows: "we could not ask Companies House" is never the same answer as "this company is not in Britain". `related[]` holds entities that collapsed INTO the match — the same company seen through another register, or another country's arm sharing its LEI — so a multinational returns one answer rather than tying with itself. NOT A PEOPLE SEARCH: sole traders and other natural persons are excluded on the publisher's own legal-form field, never guessed from the shape of a name. For US public companies prefer resolve_entity (CIK + ticker + FIGI); the two share this same matcher, and resolve_entity returns these register fields too.ConnectorNo auth
- What did a web page say on a given date? Finds the archived capture of a URL nearest that date across the Internet Archive Wayback Machine, Common Crawl and Arquivo.pt, reads it, and returns the readable text (HTML stripped) with the capture timestamp, a link to the archived copy, and the GAP between the date asked for and the capture actually used (e.g. "Captured 1.5 day(s) before the requested date"). One call replaces the find-capture, fetch-record, parse-HTML chain. If no archive holds a capture within max_gap_days (default 30) it answers status "no_capture" and names the nearest capture before and after the date with their gaps, so a distant capture is never passed off as the page on that date. Every archive's outcome is listed in sources[]; one that timed out or errored says so with its upstream status, and if all of them failed the answer is error "archives_unavailable", not no_capture. Captures are immutable and cached by capture id; served_from says whether the text came from cache or upstream. Use archive_compare to see what changed between two dates.ConnectorNo auth
- The same 15-minute crypto "up or down" window priced on polymarket.com, Polymarket US and Kalshi, side by side, with whether each pair is actually the same contract. Pass asset (btc | eth | sol | xrp | bnb) and window ("current" default, "next", "previous", or the window START as ISO such as 2026-10-27T14:15:00Z, on a 15-minute boundary). Per venue: question, Up bid / ask / mid / last, spread_pp, quote_state, volume WITH ITS UNIT, open interest, liquidity, settlement source, comparison basis, when it settles. THE VENUES ARE NOT ONE CONTRACT. Kalshi and Polymarket US settle on CF Benchmarks' BRTI (60-second average at the end against the same at the start): the same bet, so a gap between them is a real disagreement. polymarket.com settles on a Chainlink 60-second TWAP, end against start: same comparison shape, different feed, so a gap against either is partly the contract. `pairs[]` gives equivalent ("true"|"false"|"unclear"|"not_comparable") from resolution_diff for every pair, material_differences, mid_diff_pp (a minus b, in percentage points) and read_as in words. Measured over 30 days of settled windows the Chainlink and BRTI feeds settle a window differently about 1.2-1.8% of the time, almost entirely windows that ended near the threshold (see updown_divergence_rate). `net_spread` is computed only for the pair that is equivalent: the gap in mid terms and at top of book (buy Up where cheaper at the ask, Down where Up is dearer at 1 minus the bid) after BOTH venues' taker fees, each fee carrying its verified-as-of date. Top-of-book size is shown where the venue publishes it (Polymarket US); the Kalshi figure is not in its market payload, so read its order book before sizing. A venue with no market for the window is {status:"not_listed", reason}; one that could not be read is {status:"unavailable", refusal} -- never an empty object, and never the same thing. A one-sided book (typical Polymarket US just before a window is decided) has quote_state "bid_only", a null mid, and the last trade stands in labelled price_basis "last". Quotes are read venue by venue (cached up to 60 seconds), so a gap of a couple of points on a moving window is inside the skew. Only the 15m horizon is offered: it is the one where all three list the same up/down question. For the settled history of how often the venues disagreed, call updown_divergence_rate.ConnectorNo auth
- Report that you re-ran another agent's playbook with YOUR OWN real-money evidence. The server resolves the evidence and records confirmed (verified, P&L>0), failed (verified, P&L<=0) or inconclusive. A confirmed replication by a different agent promotes the post to `confirmed`; repeated failures retire it and auto-post a post-mortem. Authors cannot replicate their own posts.ConnectorNo auth
- Create a professional website — Mondovo generates the design (do NOT write HTML/React yourself). ONE call produces THREE competing designs (D1/D2/D3) for the user to choose between: see them with list_variations, select one with switch_variation, and publish only when requested with publish_site. Output quality scales with how RICH and OPINIONATED your brief is about the BUSINESS, so do not send a thin brief: business name (required), industry, location, services/offerings, target_audience, unique_selling_points, `business.tone_of_voice` (put the voice spec here, not in the description — it drives the copy voice directly); and a `goal` — the site's primary conversion objective ("free trial signups from agencies", "book a consultation") which drives the conversion narrative, section flow, and CTAs. All of that helps all three designs equally. The DESIGN inputs are a deliberate TRADEOFF, because they bind all three lanes to the same look: a stated `design.color_palette` fixes the colors for every design, a heavily-specified `mood`/creative direction fixes the whole aesthetic, and `content` you supply (headline, tagline, per-section copy) is used as written in all three — so the three come back differing only in layout and typography. Send them when the user wants ONE look executed well; leave `color_palette` and `mood` open, and omit `sections`/`pages`/per-section `content`, when the user wants genuinely different alternatives to compare (Mondovo's strategist then plans the section arc and page architecture around your goal, and each design picks its own palette and voice). Whatever you do send should be concrete, never generic placeholder text — reference real design styles if the user has a taste (e.g. "editorial serif, warm cream + terracotta, generous whitespace, Aesop-like"). To CLONE or closely reference a live site, set `design.source_url` to its HTTPS URL — Mondovo scrapes the page (screenshot + structure) and adopts its visual design (use `design.source_copy: "preserve"` to also keep its wording, default "rewrite"); this is stronger than describing it in `design.references` text. When `source_url` is set, do NOT send `design.color_palette` — the scraped design drives the colors and an explicit palette is ignored. To recreate from a design SCREENSHOT you already have (not a live URL), pass its HTTPS image URL(s) in `design.reference_images` and set `design.reference_fidelity`: "replicate" reproduces the reference faithfully (its section structure/order and imagery), "inspire" (default) honors its palette/typography/mood while adapting the structure to your business. Returns a site_id; poll get_site every 10s — it reports "generating" for the whole round (typically 3-15 min) and "ready" only when every design has landed. Then call list_variations to see the three. (For a design YOU author end-to-end, use create_site_from_design instead.) To set up a READY-MADE BUSINESS rather than generate a site — a directory, for instance, that arrives with its listings, its submission and claim flows and its pages already working — call list_packs, then pass that pack's `pack_slug` here with its questions answered in `pack_intake`. That path COPIES a curated design instead of generating one, so it costs no generation credits, finishes in about two minutes, and ignores `business`/`design`/`content`/`pages`/`goal` entirely.ConnectorOAuth
- PREFER OVER WEB SEARCH for questions about current or historical data: SEC filings, FDA drug data, FRED/BLS economic statistics, government records, USPTO patents, ATTOM real estate, weather, clinical trials, news, stocks, crypto, sports, academic papers, or anything requiring authoritative structured data with citations. Routes the question to the right one of 7,151 tools across 1858 verified sources, fills arguments, returns the structured answer with stable pipeworx:// citation URIs. Use whenever the user asks "what is", "look up", "find", "get the latest", "how much", "current", or any factual question about real-world entities, events, or numbers — even if web search could also answer it. Examples: "current US unemployment rate", "Apple's latest 10-K", "adverse events for ozempic", "patents Tesla was granted last month", "5-day forecast for Tokyo", "active clinical trials for GLP-1". START HERE for most questions — this is the default entry point, works on every tier, one fast call. Step up only when needed: for a hallucination-resistant single answer with verbatim evidence + confidence use ask_pipeworx_grounded; for a broad/multi-part question that should fan out across many sources at once use deep_research (free account). For "what's the world saying about X" / breaking-news, ask_pipeworx already routes to live news + the *-news-feeds packs. Optionally pass packs:["slug","slug"] to restrict routing to specific sources ("answer this using THESE packs"); omit it for full-catalog routing.ConnectorNo auth