"A local-only document search tool" matching MCP tools:

• get-space-documentA

  docuwriterA

  Use this when the user wants to read the full markdown content of a specific Space document/page after search or listing. Read-only: returns the selected document without changing content. Requires the document ID from list-space-documents, search-space-documents, or global-search.

  ConnectorOAuth

• getsign_report_issueA

  GetSign MCPA

  Report a GetSign problem so the team can debug it. Never sends document content. Call this when a GetSign tool failed or did the wrong thing and the user wants it looked at. GetSign's server sees only tool calls, never the chat, so this is the only way what the user was trying to do reaches us. description: what the user wanted and what actually happened, in your own words. Do not paste document text, signer emails, or tokens. tool: the GetSign tool that misbehaved, if one did. reference: the reference / support_id from the failing tool result, if present — it links this report to the exact server-side request.

  ConnectorOAuth

• upshift_site_auditA

  UpshiftA

  Fetch a live website and report how legible it is to search engines and AI assistants: structured data (JSON-LD), llms.txt, robots.txt, title and meta description, headings, image alt text, HTTPS and document weight. Returns findings with severity and a 0-100 score. Use this to check a real site before recommending changes to it. Makes one outbound request to the URL given, so it only sees what a fetch can establish — never rendering or browser metrics. Refuses private, loopback and link-local addresses, and reports a page as unreachable rather than scoring the error body.

  ConnectorNo auth

• update_datafileA

  TessryxA

  Update a datafile. This is a partial update at the FIELD level: only the arguments you provide change. Provide json to replace the WHOLE document (its property order is preserved verbatim); omit it to change only metadata. To change PART of an existing document, prefer patch_datafile — it sends just the edit and cannot lose a concurrent write. Use this tool when you are replacing the whole document anyway; both preserve property order. Sending json here means get_datafile, change the object you got back, then send it all — pass expected_content_sha256 with that round trip so a change you did not see is reported instead of overwritten. Republish to push the change to the CDN.

  Connector

  Destructive

  OAuth

• create_update_document_adA

  Metadata MCP ConnectorA

  Create a new or update an existing LinkedIn Document Ad on the Metadata platform. A LinkedIn Document Ad promotes a multi-page document (a PDF — e.g. a whitepaper, ebook, report, or guide) natively in the LinkedIn feed. Members preview the first few pages inline, then unlock the full document (typically gated behind a lead form). It is LinkedIn-only. IMPORTANT: This tool is ONLY for LinkedIn Document Ads (channelType=LINKEDIN, adType=DOCUMENT). For IMAGE/GIF ads use create_update_image_ad; for VIDEO ads use create_update_video_ad; for CONVO use create_update_convo_ad; for Sponsored Messaging use create_update_linkedin_message_ad. BEHAVIOR: - If `id` is NOT provided -> creates a new Document Ad. - If `id` IS provided -> updates the existing Document Ad with that ID. CREATIVE WORKFLOW (MUST DO BEFORE CALLING): 1. The document must already exist in the creative library as a DOCUMENT asset. Find it with `search_library_creatives_by_name(contentTypes="DOCUMENT")`, or confirm a specific id with `fetch_creative_details`. 2. Pass that asset's integer id as `libraryId`. It MUST be a DOCUMENT-type creative — an image or video id will be rejected by LinkedIn. Unlike IMAGE/VIDEO ads there is NO display URL: the gated document is the destination, so no link field is accepted. CHARACTER LIMITS (mirror the LinkedIn ad limits the platform UI enforces): - name <= 50 chars (ad name in the library) - headline <= 200 chars (the headline shown with the document) - text <= 3000 chars (the introductory text) INPUT PARAMETERS: - id (optional): Existing Document Ad ID. Provide to UPDATE; omit to CREATE. - name (required, <= 50): Ad name in the library. - libraryId (required): Integer id of the DOCUMENT creative in the library (contentType=DOCUMENT). NOT an image/video id. - headline (required, <= 200): Headline shown with the document. - text (required, <= 3000): Introductory text shown with the document. - ctaType (optional, default UNLOCK_FULL_DOCUMENT): the call-to-action. UNLOCK_FULL_DOCUMENT is the document-gate CTA. - maxPreviewPages (optional, default 1): how many pages of the document are previewable before the unlock gate. Positive integer. - completionStatus (optional, default DRAFT): "DRAFT" or "COMPLETED". Use COMPLETED only when every required field is final. WHEN TO USE: - User asks to "create a LinkedIn Document Ad" / "whitepaper ad" / "promote a PDF / ebook / report on LinkedIn". - User wants to update an existing Document Ad. WHEN NOT TO USE: - Feed image / GIF ad -> create_update_image_ad - Feed video ad -> create_update_video_ad - Branching conversation flow -> create_update_convo_ad - One-shot Sponsored Message -> create_update_linkedin_message_ad EXAMPLE USAGE (Create): create_update_document_ad( name="Q3_Whitepaper_LI_Document", libraryId=15791, headline="The 2026 State of B2B Marketing", text="Download our latest research on pipeline attribution.", maxPreviewPages=2, ) EXAMPLE USAGE (Update): create_update_document_ad( id=29144, name="Q3_Whitepaper_LI_Document_v2", libraryId=15791, headline="The 2026 State of B2B Marketing (Updated)", text="Now with fresh benchmarks.", ) COMMON MISTAKES: - Passing an IMAGE/VIDEO libraryId — the asset must be a DOCUMENT (use search_library_creatives_by_name(contentTypes="DOCUMENT")). - Passing a display / landing URL — Document Ads have no link field. - Using this tool for a non-LinkedIn channel — Document Ads are LinkedIn-only.

  Connector

  Destructive

  API key

• qbSendEmailA

  DeepLedger — AI staff accountant for QuickBooksA

  Email a sales document (Invoice, Estimate, SalesReceipt, CreditMemo, RefundReceipt) to the customer through QuickBooks. The ONLY tool that emails a customer: recording never sends, and emailStatus:"NeedToSend" on qbInvoice/qbEstimate only queues the document for a human to send. An email cannot be unsent: confirm with the user first, never send to test, and never re-send to be sure (every call delivers again). Needs documentType and documentId. Delivery is confirmed only by delivered:true and deliveryTime, not by emailStatus. Voided documents and documents with no address are refused. ERRORS: getGuide(guideType:"error_recovery", tool:"qbSendEmail").

  Connector

  Destructive

  OAuth

• edit_documentA

  ProductNowA

  Edit an existing ProductNow document. This is the only tool that changes a document's content. It does not apply your edit directly: it hands `message` to the document's own editing agent, which reads the document and makes the change. Write `message` as instructions to that agent, not as the new document text. Name the section, say exactly what to add, replace, or remove, and quote text verbatim when the wording must be preserved. Include everything the agent needs, since it cannot see this conversation and will not ask follow-up questions. Returns before the edit finishes; poll get_status until status is idle (it stays running while any edit you sent is queued or in progress), then read the document to confirm the result. Do not use create_document — that creates a new document. Do not use rename — that only changes a name. Do not use reply_to_thread — that only adds discussion and never changes content.

  Connector

  Destructive

  OAuth

• get_filingA

  HKEx FilingsA

  Download one HKEx document by its URL and read, page through, or search its text. ``link`` must be an HKEx document URL from search_filings (host ``www1.hkexnews.hk``); any other host is rejected. With ``extract=True`` (default) the response includes up to 30 ``tables`` and a window of extracted ``document_text``: ``max_chars`` characters starting at ``offset``. Long reports return ``next_offset``; call again with it to read the next window. Pass ``query`` to search the whole document instead: the response then lists up to 20 case-insensitive matches with their offsets and surrounding text, plus ``match_count`` — use a match offset to read that part. Set ``extract=False`` for size and content type only.

  ConnectorNo auth

• axint.registry.searchA

  AxintA

  Search the Axint Registry for already-published packages that match a natural-language query. Use this BEFORE calling axint.feature or axint.compile so the agent can install an existing package instead of regenerating Swift the community has already shipped. Use: use before generating code to find reusable packages; not for validating local Swift. Inputs: query drives ranking; kind and platform narrow results without changing the registry source. Effects: read-only local registry search using AXINT_REGISTRY_PATH or sibling checkout; no network by default.

  ConnectorNo auth

• sieve_dataroom_addA

  SieveA

  Add a document to a deal's data room. Creates the deal if needed. This is the primary way to get documents into Sieve for screening. Upload a pitch deck, financials, or any document -- then call sieve_screen to analyze everything in the data room. Provide company_name to create a new deal (or find existing), or deal_id to add to an existing deal. Provide exactly one content source: file_path (local file), text (raw text/markdown), or url (fetch from URL). Args: title: Document title (e.g. "Pitch Deck Q1 2026"). company_name: Company name -- creates deal if new, finds existing if not. deal_id: Add to an existing deal (from sieve_deals or previous sieve_dataroom_add). website_url: Company website URL (used when creating a new deal). document_type: Type: 'pitch_deck', 'financials', 'legal', or 'other'. file_path: Path to a local file (PDF, DOCX, XLSX). The tool reads and uploads it. text: Raw text or markdown content (alternative to file). url: URL to fetch document from (alternative to file).

  ConnectorNo auth

• searchA

  Melvea Local Honey DiscoveryA

  Search the Melvea local honey directory by free-text query and return matching producers as a list of results (id, title, url). Designed for ChatGPT Deep Research and Company Knowledge. Use for any local-honey discovery query that names or implies a place; the tool parses place and varietal from the query. Returns an honest empty list when nothing matches — never fabricate. Pair with fetch to retrieve full producer detail.

  ConnectorNo auth

• dd_company_filing_documentA

  UK Business Tools - LedgerhallB

  Resolve a filing's document_metadata link to its authoritative source document. Returns a resource_link (never embedded bytes, never base64) pointing at a company-document:// MCP resource — fetch it via resources/read to get the actual PDF. This tool only reads metadata (category, pages, available content types, byte size); it never downloads the document itself. Use company_filing_history first to find a filing's document_metadata URL. Requires a resource-capable MCP client to retrieve the actual bytes — a tool-only client can see this result's metadata (company, category, page count, size) but cannot obtain the file through this tool call alone.

  ConnectorNo auth

• search_gmailA

  Dayze — Life ContextB

  Search one of the authenticated user's connected Gmail inboxes (read-only). Reuses Integrations hub Gmail (service_connections provider=gmail). Optional account/email selects a mailbox when multiple are connected — call list_gmail_accounts first. Prefer Gmail operators: from:, subject:, is:unread, newer_than:7d, has:attachment. For clock-time constraints, pass after/before as account-local civil datetimes; the tool converts them to precise UTC instants. Returns id, subject, from, canonical date, account-local date_local, time_zone, and snippet only (no body), newest first (newest_first: true). For the latest / most recent email from someone, query from:<sender> alone, without quoted topic phrases (the newest reply often lacks them); a from: + topic search with no match from the last 7 days also searches the sender alone and marks those messages sender_only (see note). Follow up with read_gmail_message. Cannot send, archive, or delete. Never invent email contents. If Gmail is not connected, connect at /integrations. Requires mail.read on an API key or non-ChatGPT OAuth connection. ChatGPT profiles do not grant mail.read or list this tool. Share tokens cannot call this tool. ($0.10; API key required)

  ConnectorNo auth

• siteborne_get_service_healthA

  SITEBORNE Utility NetworkA

  Report MCP server readiness and production-enable status for each SITEBORNE service. Use when: an agent must check protocol availability, tool count, or whether a service is currently production-enabled before selecting a paid tool. Do not use when: a quote is needed (use siteborne_get_quote) or company, web, document, or agent-output evidence work is required (use the corresponding SITEBORNE service tool). Behavior: read-only and credential-independent; it does not perform paid evidence work, create quotes, verify payment, call providers, write service state, create Workflows, or settle. Economics: free of charge; it never issues a payment challenge. Failure: it depends on no provider, so it does not fail because a provider is unavailable. Returns: the server and protocol versions plus truthful local, production, and external-publication status for all four evidence services.

  ConnectorNo auth

• create_documentA

  ProductNowA

  Create a NEW ProductNow document and generate its content with AI. Use this to persist knowledge the user wants remembered or shared across sessions and teammates — decisions, specs, RFCs, plans, meeting outcomes. Search with search first so you don't duplicate an existing document; to change an existing document use edit_document instead. File the document in an existing folder. Search results include each document's folderId; pass the folderId of the closest related document. If search does not show a fit, fetch_folder the root and walk folders until you have a folderId. Leave folderId off only when the user explicitly asked for the warehouse root. If no folder fits, call create_folder under the closest existing folder and pass that new folderId. Content is generated from `prompt` (the full outline: numbered sections, their titles, and what each should say) and `context` (verbatim source material the user supplied or explicitly authorized for this ProductNow document). Generation paraphrases: if exact wording must survive, put it in `context` and say so in `prompt`. Pass `templateId` only when you already have one. Returns the documentId and URL; generation continues after the call returns.

  ConnectorOAuth

• rememberA

  ProductNowA

  Store one piece of information in ProductNow when the user says 'remember this', 'save this', or 'note that…' about their company, product, customers, or team. Pass the fact as `content`; ProductNow decides where it belongs — it searches the warehouse, edits the existing document that covers the topic, or creates a new one in the right folder. ProductNow handles search, folder selection, and document choice within the background job, so separate search, fetch, or placement is not needed. Returns success immediately; a success response means the job was queued and processing continues in the background. Polling is not supported. Use edit_document or create_document instead only when the user already named the exact document or outlined a whole new document.

  Connector

  Destructive

  OAuth

• batch_documentsA

  DocsMintA

  Use for multiple explicitly selected documents that need the same action; for one document prefer the matching single-document tool. Accepts 1–25 IDs for automation compatibility. Supports move, set_category, add_tag, remove_tag, trash, restore, or refresh_index. Processes sequentially through the existing API with a permission check per document and a per-ID result; partial success is possible. Stops after authentication or rate-limit failure. Never expands a search result or permanently purges documents; version restore is separate.

  Connector

  Destructive

  API key

• tool_schemaA

  foundryA

  Read the full JSON Schema of any OODS Foundry tool at the pinned version, as the package ships it: name is any of its tools, local-only ones included, and kind is input (the default) or output. Returns the schema, its file in the package, that file’s SHA-256 and its URL on this site. Use it before calling a tool over local stdio, or to learn what a tool returns: these hosted tools declare no output schema, and tools/list here shows narrower inputs than local stdio accepts.

  ConnectorNo auth

• works_public_eligibilityA

  WORKS Public VerifierA

  Fast deterministic preflight for tool-only clients. Call this before any other WORKS tool when eligibility is uncertain, especially for mutable or abbreviated refs, local or private repositories, and build, test, runtime, deployment, or production claims. It does not download a repository or persist data. If eligible is false, stop without calling verification.

  ConnectorNo auth

• SearchDocumentA

  equiblesA

  Search one SEC filing or earnings-call transcript by document ID. semantic mode uses hybrid relevance and returns excerpts in document order with approximate line numbers. exact mode performs a literal case-insensitive substring match and returns precise matching lines. Get document IDs from SearchDocuments or ListFilings; use ReadDocumentLines for surrounding text.

  ConnectorNo auth

• search_linkedin_peopleA

  Salesbot - B2B outreach LinkedIn MCP Server with CRM and emailA

  Default tool for finding people on LinkedIn from criteria (company, role, department). With a ready search URL use search_linkedin_navigator; for one known profile URL use upsert_linkedin_contact. Search LinkedIn people. RECOMMENDED FOR PROSPECTING: set decisionMakers:true, provide company, departments and limit 20-30. Salesbot resolves the company, performs ONE company-scoped provider search (Sales Navigator also applies seniority), then ranks the returned senior employees locally by department. This is broader and safer than retrying exact titles. Use title/titles only when an exact role is required. EXISTING CONNECTIONS reads only the local cache and consumes zero search quota. The workspace setting selects Standard/Classic or Sales Navigator automatically. Respect retry_after; never immediately retry a protected or timed-out request. On LINKEDIN_PROVIDER_TIMEOUT use search_google_xray, then retry LinkedIn only after the stated delay.

  ConnectorAPI key