"Document analysis tools for legal documents and evidence" matching MCP tools:

• search_content

  flevy

  Search Flevy's marketplace of consulting frameworks, PowerPoint templates, Excel financial models, business toolkits, and management case studies. Use this whenever a user needs a best-practice framework, methodology, template, financial model, or real-world case example on any business or management topic (strategy, digital transformation, supply chain, pricing, operational excellence, M&A, etc.). Returns up to 10 relevance-ranked recommendations across two content types: "document" (premium documents authored by management consultants) and "case_study" (management case studies). ALWAYS include each recommended item's url as a clickable link when you mention it in your reply — never reference a document without its link, because the link is the only way the user can open it. Each result carries a content_id for get_content_details. Filters: topic (single, or "topics" for documents covering ALL of several topics), author (list more documents from an author seen in results), filetype (including tier1_consulting_deck for McKinsey-style strategy decks), content_type. Topic-filtered responses also list related_topics to pivot to. Provide at least one of query, topic(s), or author; use list_topics to map user phrasing to a canonical topic.

  Connector

• search

  Islam West Africa Collection (IWAC)

  Search the Islam West Africa Collection across newspaper articles, Islamic publications, archival documents, academic references, audiovisual recordings, photographs, and the authority index (persons/places/organisations/events/subjects). Pass ONE concept or name — e.g. 'Tijaniyya', 'laïcité', 'Sheikh Gumi', 'pèlerinage'. Matching is accent- and case-insensitive; a multi-word query requires every word to appear somewhere in the item, so prefer a single concept per call. Write query strings and concept keywords in French for press/publication/document/index discovery even when the user's report language is not French. Academic references are multilingual, so try French and English title/abstract terms when relevant; metadata/filter labels remain French. Use the French transliteration of Islamic terms (Tabaski not 'Eid al-Adha', charia not 'sharia', Maouloud not 'Mawlid'). Returns {results:[{id,title,url,category}], ranking}; each result's `category` names its subset and the `ranking` field documents the ordering. Pass an id to `fetch` to read the full text. For filtered queries (by country, date, or newspaper) use the search_* tools instead.

  Connector

• verify_identity

  Stipple — Document Verification & Extraction

  Run an Australian identity check over a SET of identity documents. A vision model reads each document (which ID it is, which fields it shows — name/photo/address/signature — and its issue date); a deterministic engine then tallies them against a scheme and reports whether identity is established, and exactly what's still missing if not. USE THIS WHEN someone needs to verify a person's identity from their documents — KYC / onboarding / "do these documents satisfy the 100-point check?" Pass ALL the person's documents together (a passport alone is 70 points; the check needs >= 100). `documents` is a list, each item ONE of: {"url": "https://..."} (public link, fetched server-side) or {"bytes_b64": "...", "filename": "passport.pdf"} (inline). Up to 10. `scheme`: "afp_100_point" (points, default) or "austrac_safe_harbour" (category combinations). Returns `{established, points/target or satisfied_path, documents[] (per-document: type, fields shown, whether it counted and why-not), reason, accepts, ...}`. This is identity COVERAGE, not a forgery judgment — run verify_document for authenticity. Documents are never stored. Costs 2 credit(s) per call.

  Connector

• check_pack

  Stipple — Document Verification & Extraction

  Check whether a SET of documents satisfies a checklist — completeness, cheaply. USE THIS WHEN you have an application / onboarding pack and need "do we have the required documents, and what's still missing?" Each document is CLASSIFIED (one cheap page-1 read — never full field extraction or multi-page), then matched against the checklist's required slots. (For "is a document genuine?" use verify_document; to identify ONE document use extract_fields with options={"classify": true}; for the identity gate use verify_identity.) Define the checklist ONE of two ways: - `scheme`: a named preset — "income_proof", "lending_prequal", "rental_application". - `requirements`: an ad-hoc checklist — a list of document-type names like ["payslip","bank_statement"], or objects {"key":..., "accepts":[types], "optional":bool}. `documents` is a list (up to 12), each ONE of: {"url": "https://..."} (public link, fetched server-side) or {"bytes_b64": "...", "filename": "statement.pdf"} (inline). Returns `{complete, slots[] (key, satisfied, matched), missing[], documents[] (filename, classified_type), unmatched_documents[]}`. COVERAGE, not approval — that the right document TYPES are present, NOT that any is genuine (run verify_document) or that an application is approved. Documents are never stored. Costs 3 credit(s) per call.

  Connector

• document.check_claims

  api

  Verify a list of factual claims against document text. Uses a quality AI model with citation-level evidence. Use after document.extract_text or url.extract when you need to validate specific factual assertions. For open-ended questions about a document, use url.qa instead. For multi-document investigation, use collection.ask. Typical workflow: document.extract_text/url.extract → document.check_claims. Returns: { claims: [{ claim, status: "supported"|"contradicted"|"not_found", evidence: { quote, paragraphs[] }, confidence: "high"|"medium"|"low" }], truncated: boolean } Example prompts: - "Check whether this contract mentions a liability cap of $1M." - "Verify these claims against the document: [claims list]." - "Does the report actually say revenue grew 23%?"

  Connector

• analyze_title_documents

  Curagent Title Risk

  Analyzes Florida real estate title documents and returns a structured risk report: a risk score (0-100, higher is safer) and level, findings with verbatim evidence from the documents and guidance on how to cure each one, and the Schedule B-I requirements extracted from any title commitment in the package. Accepts one or more PDF documents as base64 strings (deed, title commitment, mortgage, closing disclosure, survey, payoff letter, HOA estoppel, etc.), including a single PDF containing a whole closing package, which is split into its constituent instruments. Submitting several documents together also enables cross-document checks for contradictions in parcel ID, address, and party names. Uses one of your 3 free analyses (sandbox tier) or 1 credit (paid tiers). Florida properties only — call check_coverage first to confirm scope, and get_credit_balance to confirm available credits.

  Connector

• document.reprocess

  FlexOrch MCP

  Re-queue an already-uploaded document through the pipeline. Use this to re-run extraction/PII detection/quality scoring on a document you've already processed — e.g. after a document_type_hint change, without downloading and re-uploading the original file. Returns immediately with a job_id — poll with job.status the same way as after document.process. Only works for documents whose original file is still stored on the server (locally uploaded, not connector-sourced). If the file is no longer available, re-upload it with document.process instead. Args: document_id: ID of a previously processed document. pipeline_config: Optional pipeline config overrides, e.g. {"document_type_hint": "invoice"}.

  Connector

• update_rule

  Sweeppea MCP

  Update an existing official rules document. Use fetch_rules first to get the rules_token. LEGAL BYPASS WARNING: Updating document_content replaces a legally valid document with free-form HTML that bypasses every safeguard of create_rules_wizard (AMOE, 21+ alcohol gate, COPPA, state eligibility). NEVER compose, draft, or modify the legal language yourself. Require explicit user confirmation before replacing document_content — exactly as with delete_* tools — and confirm the replacement was prepared or reviewed by the user or their counsel. UPDATABLE FIELDS: Only these fields can be modified: title, document_content, abbreviated_rules_shopify. NOT UPDATABLE: sweepstakes association, primary status, creation date, and any other field NOT listed above cannot be changed after creation. Do NOT tell the user they can update fields that are not supported by this endpoint. If they ask to change something not updatable, explain it cannot be modified after creation. # update_rule ## When to use Update an existing official rules document. Use fetch_rules first to get the rules_token. LEGAL BYPASS WARNING: Updating document_content replaces a legally valid document with free-form HTML that bypasses every safeguard of create_rules_wizard (AMOE, 21+ alcohol gate, COPPA, state eligibility). NEVER compose, draft, or modify the legal language yourself. Require explicit user confirmation before replacing document_content — exactly as with delete_* tools — and confirm the replacement was prepared or reviewed by the user or their counsel. UPDATABLE FIELDS: Only these fields can be modified: title, document_content, abbreviated_rules_shopify. NOT UPDATABLE: sweepstakes association, primary status, creation date, and any other field NOT listed above cannot be changed after creation. Do NOT tell the user they can update fields that are not supported by this endpoint. If they ask to change something not updatable, explain it cannot be modified after creation. ## Pre-calls required 1. `fetch_sweepstakes` if the user gave you a sweepstakes name instead of a token ## Parameters to validate before calling - `sweepstakes_token` (string, required) — The sweepstakes token (UUID format) - `rules_token` (string, required) — The rules token to update (UUID format) - `title` (string, optional) — New title for the rules document (max 100 characters) - `document_content` (string, optional) — New HTML content for the rules (max 1,000,000 characters) - `abbreviated_rules_shopify` (string, optional) — Abbreviated rules for Shopify integration (max 1,000,000 characters) ## Notes - Replacing `document_content` swaps a legally valid document for free-form HTML that bypasses the wizard safeguards — require explicit user confirmation first, like delete_* tools - NEVER compose or modify the legal language yourself

  Connector

• get_legal_documents

  Molt2Meet

  Get all active legal documents an agent must accept on registration. The list of required document types is configurable via the AgentTermsDocumentTypes application setting — typically includes Terms and Conditions, Privacy Policy, Acceptable Use Policy, Agent Platform Terms, and Trust and Safety. Each document includes its type reference, name, version, effective date, and full markdown content. Call this before register_agent so you know what the agent is accepting when setting acceptedTerms=true. No authentication required.

  Connector

• word_append

  local-mcp

  Appends text to the end of an existing Word (.docx) document at `path`, preserving the document's existing content and formatting. Requires confirm=true — called without it, returns a preview instead of modifying the file. Same file-access rules as word_create (Desktop/Documents/Downloads may need a Files-and-Folders grant). Returns {appended, chars_appended, path}. To create a new document use word_create; to read one use word_read.

  Connector

• ris_track_changes

  ris-austria-mcp-server

  Track every document added or changed in one RIS application within an exact date window (changed_from/changed_to), optionally including deletions (include_deleted) — the delta-sync and monitoring primitive for mirrors and watchers, and the only surface that reports removals. Unlike the search tools’ coarse, additive-only changed_since intervals, this is exact-dated and deletion-aware. application accepts every RIS application code listed by ris_list_reference topic applications, including BrKons, LrKons, Dsk, and BgblAuth. Each changed document comes back in a compact cross-class record — document_number (for ris_get_document), title, dates, binding_status, and rendition URLs — plus its last-changed date; removed documents come back as deleted records with a deletion timestamp. One application per call; page explicitly for large windows.

  Connector

• collection.ask

  DocImprint

  Answer a question using RAG over a document collection. Retrieves relevant chunks then synthesizes a cited answer with source attribution. Use when you need a direct answer grounded in your collection documents. For raw matching chunks (without synthesis), use collection.search instead. For single-document Q&A, use url.qa instead. PREREQUISITE: Collection must be populated via collection.add_document and indexed before results appear. Returns: { answer: string, sources: [{ bundle_id, chunk_id }], retrieval: [{ bundle_id, chunk_id, text, score }] } Example prompts: - "What are the key terms of the service agreement in my collection?" - "Based on my due diligence docs, what are the main risks?" - "Answer this question using all documents in the Q4 Contracts collection."

  Connector

• validate_portfolio

  aibvf-mcp

  Check whether a supplied AI BVF v1.0 portfolio document has the shape the portfolio tools require, before scoring, sequencing, storing or sharing it. CALL THIS when the document came from a file, another system or hand-built JSON and its structure is uncertain. It checks required fields, taxonomy values and 0–100 pillar ranges only; it does not judge the evidence or calculate a verdict. Pillars may be bare numbers or { value, confidence } objects, both are valid. Use assemble_portfolio when the user has a list of initiatives in conversation and needs the document built for them, score_portfolio when the document is already ready for verdicts, and sequence_portfolio only after its initiatives are scoreable. Returns valid=true or one error per failing JSON path. Pure deterministic validation — no network, auth, or side effects.

  Connector

• stage_document

  Taokeh MCP server

  Bring ONE already-issued historical document (a sales invoice, or a supplier bill) across from the system this workspace is migrating FROM. This is the BULK migration door and it is NOT create_invoice_draft: nothing is posted, nothing is drafted for individual approval, and no approval card is raised per document. Staged documents group into monthly batches the owner reviews and approves together at /switch/documents. Use it ONLY for documents that were genuinely issued in the old system (or written in a paper book) — a NEW document belongs in create_invoice_draft / create_bill_draft. The lane must be OPEN (the owner turns it on at /switch → "Bring over your documents") and it can only be open before the books are locked; if it is closed the call is refused with instructions. IDEMPOTENCY IS ON sourceDocId: staging the same sourceDocId again UPDATES that pending row instead of filing a second one. Once the owner has ENTERED it, or deliberately LEFT IT OUT, nothing you send can change it, re-book it or bring it back — only the owner can. So re-running your whole export is always safe. THE CUTOVER: bring over only documents dated AFTER the workspace's accounting start date. Anything dated on or before it is already carried by the owner's opening balances, and Taokeh will flag it and refuse to enter it — staging those wastes both our time, so filter them out of your export if you can. ORDER MATTERS: stage the SUPPLIER BILLS for a period before the sales invoices for it, because Taokeh works out cost of sales from the stock that was bought. The server re-computes every quantity and the grand total from the lines (with SST); your own printed total goes in sourceTotal and is used ONLY to show the owner a tie against the server's figure. Anything the server cannot settle — a product that does not resolve, a customer name matching several contacts, a printed total that disagrees — is staged anyway, FLAGGED, and held out of bulk approve for the owner to open individually. The document keeps its ORIGINAL number (reference) and its ORIGINAL date, and posts marked as historical so Taokeh never e-invoices or chases it. PAPER: if you read this document off a photo or a scan, ATTACH IT — request_attachment_upload, PUT the bytes, pass attachmentToken here. The owner then reviews your figures beside the actual slip on that document's own screen, edits anything you misread, and approves it there; the original lands on the posted document. Without it they are approving your arithmetic on your word alone. Show your per-line working, and mark any balancing/catch-all line residual:true — a residual line carrying a material share of the total is flagged for the owner, because that is exactly where a misread hides.

  Connector

• read_document

  ocr

  Read a document (PDF or image) from a URL and return its contents as markdown (tables preserved) or plain text. Costs $0.00075 per page, billed to the PennyOCR account; the response includes the exact cost_usd and per-page citations. Use estimate_cost first for big documents. Supports page ranges and hard spend caps.

  Connector

• get_evidence_document

  Plumbline — a trust layer for LLM answers

  Show ONE retrieved evidence document behind an answer you already received, addressed by that answer's correlation_id plus a document_id from its evidence_documents references. Returns the full stored document (title, body, metadata, embedding_text) with the retrieval rank and scores the answer recorded; never the raw embedding vector. Only documents the addressed answer actually recorded resolve: there is no fetch-by-id in general and no way to browse the store. Requires the persisted compliance log and the same session that produced the answer. Absent from the no-auth public demo.

  Connector

• search

  FirmTape - SPX dealer positioning

  Search FirmTape for documents about SPX dealer positioning: the explainer pages, the dated research measurements, and the archive of finished trading sessions. Returns ids to pass to `fetch`. A date in the query ("2026-08-24", "August 24 2026", "August 2026") finds the session or sessions for it. Clients that can call the specific tools should prefer list_sessions / get_session / get_levels / get_level_history instead — those return structured numbers rather than documents. Not for: fetching a document you already have the id for (`fetch`) or any measurement you can name a date for. Limits: FirmTape's own public pages and finished sessions only, ranked by keyword — it searches no other site.

  Connector

• search_documents

  OpenPublica

  Search adopted plan and budget documents, returning page-cited excerpts. Each hit's `text` is a short passage around the match, not the full page — call read_document_pages with the hit's `first_page` to read it in context. Every hit carries `first_page`/`last_page` and `source_url`. Quote those when citing: the page number is what makes the claim checkable against the city's own copy. This is word matching, not semantic search — try the terms a plan would actually use. Only currently-served documents are searchable, so an empty result is not evidence the government has no such policy; check list_government_documents for what is held and what could not be read. A hit with `document_complete: false` comes from a document with transcription findings. Verify it against `source_url` before quoting a number from it.

  Connector

• search

  Dados Abertos Senado BR MCP

  Searches the Brazilian Federal Senate open data (senators in office and active committees of the Senate and the National Congress) catalog and returns up to 10 matching documents as { id, title, url }, ordered by relevance (an empty list means nothing matched). This tool exists for the OpenAI Deep Research contract: ChatGPT deep research, company knowledge and research workflows over the Responses API require exactly the tools `search` and `fetch`. Pass one of the returned ids to `fetch` to read the document. For direct questions and for data (values, series, rankings) prefer the `senado_*` tools, which return the actual data with provenance — this is a catalog index, not a data query. Query: natural language or keywords, Portuguese or English; accents and case are ignored. Behavior: read-only and idempotent — the catalog comes from the public source and is cached in memory.

  Connector

• search

  UNESCO UIS — Education, Science & Culture Statistics (provenance-first)

  Searches the UNESCO UIS statistics (≈5,000 indicators: education — enrolment, completion, literacy, teachers, spending, SDG 4 —, science/R&D (SDG 9.5), culture (SDG 11.4) and demographic context) catalog and returns up to 10 matching documents as { id, title, url }, ordered by relevance (an empty list means nothing matched). This tool exists for the OpenAI Deep Research contract: ChatGPT deep research, company knowledge and research workflows over the Responses API require exactly the tools `search` and `fetch`. Pass one of the returned ids to `fetch` to read the document. For direct questions and for data (values, series, rankings) prefer the `uis_*` tools, which return the actual data with provenance — this is a catalog index, not a data query. Query: natural language or keywords, Portuguese or English; accents and case are ignored. Behavior: read-only and idempotent — the catalog comes from the public source and is cached in memory.

  Connector