- Brand
- Addy AI
- Category
- Finance
- Primary Subcategory
- Mortgages & Home Financing
Integration details
Description
This app is the best way to track your loan pipeline, track your document processing checklist, see loan details, follow up for documents and much more.
- Integration type
- Plugin
- Verification status
- Not applicable
- Platform
- ChatGPT
- Primary Subcategory
- Mortgages & Home Financing
- Secondary Subcategories
- None listed
- Brand
- Addy AI
- Access
- Account required
- First tracked
- 2026-04-10
- Tool count
- 30
- Market
- US
The Primary Subcategory used for this profile’s headline score.
Other Subcategories where the Integration is listed.
Your score is coming
ChatGPT now suggests Plugins on its own when they match a user's request.Your Plugin Discovery Score measures how often yours appears, and it will show here as soon as it’s ready.
What discovery looks like

Get alerts for Addy AI
Get updates when Addy AI’s Discoverability Score or category rank changes.
Competing in ChatGPT Mortgages & Home Financing
View Category30 tools agents can invoke
Checks for likely-duplicate borrower and loan records before records are created from a prequalification. Takes the borrower identity (firstName, lastName, email, phone) and propertyAddress, or the full answers map. Returns candidate borrowers and loans with match reasons and a recommendation to proceed or review. A matched borrower's id can be reused as useExistingBorrowerId when the records are created.
check_prequalification_duplicates
Confirms the signed-in Addy account and organization for this connection, once per connection. Until then, the first organization-scoped Addy call returns this review automatically. Without confirmed it returns a review card for the proposed organization, or, when the account has several organizations and none is chosen yet, a question asking which one to use (no organization is proposed). With organizationName or organizationId it returns the review for that organization instead. With confirmed: true and the organizationId from the card, after the user approved it (on the card or in chat), it saves the choice. The review shows the account email and organization names in plain language.
confirm_addy_context
Creates a borrower record in Addy AI. Needs the borrower's name and at least one contact method (email or phone). Without confirmed it returns a review card with the borrower details and creates nothing; with confirmed: true, after the user approved that card, it creates the borrower. This creates a borrower record only; it does not submit a loan application. An idempotencyKey makes retries safe without creating duplicates.
create_borrower
Creates a working loan file in Addy AI, like Addy's manual "Create" flow: it lands in the organization's first loan-board column (not a synthetic draft) and is tracked against a real loan type. Takes 1–5 borrowers in the borrowers array (the first entry is the primary borrower); the legacy single-borrower fields are still accepted as the primary borrower. loanType is an exact canonical loan type / checklist template ID from the organization's loan types, or a generic agency (Conventional, FHA, VA, USDA, Jumbo) together with a loanPurpose that agency offers (e.g. Purchase, Refinance, Cash-Out Refinance), which is resolved to the exact template. An ambiguous or unsupported combination, or an invented type such as "other", is rejected with an error rather than mapped to a default. Without confirmed it returns a review card with the loan summary and creates nothing; with confirmed: true, after the user approved that card, it creates the loan. The loan is never submitted as an application and never synced to an external LOS. A confirmed loan needs the same minimum as Addy's form: loan type, a positive loan amount, and the primary borrower's first name, last name, and email; the response reports missingForCompletion / isComplete when data is incomplete. An idempotencyKey makes retries safe without creating duplicate loans.
create_loan_file
Converts a reviewed prequalification into Addy records: a borrower record, a loan file, and a prequalification record. Without confirmed it returns the borrower, co-borrower, loan, and prequalification details for the loan officer to review and creates nothing; confirmed: true creates the records they approved. Five prequalification fields are required: borrower first name, last name, email, loan type, and requested loan amount. The loan type must be an exact canonical loan type from the organization's loan types, or a generic agency (conventional, fha, va, usda, jumbo) with a matching loanPurpose; an ambiguous, unsupported, or invented ("other") loan type is rejected and no records are created. Each call takes the full answer set; alternatively earlier answers go in previousAnswers and new or changed fields in answers. Creates working records only: it never submits an application and never states a lending decision. If likely duplicates are found, it returns a duplicate result without creating anything; useExistingBorrowerId reuses a borrower, and ignoreDuplicates: true proceeds anyway. The same idempotencyKey makes an exact retry safe; each separate intended loan, including property-TBD loans, takes a distinct key. On success, returns the Addy loan file link.
create_loan_from_prequalification
Explains why documents are still needed on a loan: which requirements are outstanding, why each one matters (AUS conditions, checklist tasks, stored reasons), and the next action. Only requirements that are still outstanding are returned; documents the borrower already provided are excluded. Takes a loanId, or a borrowerName or query (e.g. "Jane Doe") to find the loan.
explain_missing_documents
Retrieves an authorized loan by ID, from search results or a loan ID the user gives: borrower, property, loan amount, loan number, stage, contacts when available, and an Addy loan link. Full LOS/MISMO application data is left out by default because it can be large. When heavyDataAvailable is true and the user asks for that data, includeHeavy: true also returns a size-capped, redacted digest of it.
fetch
Finds the borrower loan file that should receive uploaded documents, by borrower name, property address, or loan number. Returns candidate loan files; when more than one matches, the loan officer picks the exact destination.
find_loan_for_upload
Generates a secure Addy-hosted upload link scoped to one loan file, so documents are uploaded into the right place. It does not email the borrower or add items to the borrower portal checklist. Audience "self" is for the loan officer to upload from their phone or computer; "borrower" is for sharing with the borrower or client (borrowerEmail authorizes them for the loan). The recipient opens the link, signs in to the Addy portal, and uploads documents, which attach to this loan file and are classified and indexed automatically. The link is created with confirmed: true, after the user approved the loan file and recipient. Access is controlled by Addy portal login and loan authorization, so expiresInHours is advisory and not a hard token expiry.
generate_upload_link
Reports the status of loan-origination actions from the audit trail (created, duplicate, failed, pending, or permission denied): whether a borrower or loan was created, a document request was sent, or an upload link was generated. Takes an actionId for a single action, or filters by loanId, action, status, and a startDate/endDate window. Scope depends on the Addy role: admins and owners can review all users in the organization (and may filter to a specific userEmail); everyone else only sees their own actions.
get_action_status
Returns the status of documents requested from a borrower on a loan file: which requested documents were received (uploaded) and which are still outstanding. Counts are loan-wide across all requested and received documents, not just the most recent request. For the five most recent tracked requests it also returns the text's delivery status, whether the email was accepted, and when the borrower opened the request in the portal; requests sent before tracking existed are not listed. Email delivery itself is not tracked, only whether the email was accepted.
get_document_request_status
Returns Addy AI's calculated processing-checklist result for one loan topic: income, credit, or assets. Covers questions such as "what is the income on Jane Doe's loan?", "what are the credit findings?", or "what assets or down payment were verified?", including W-2, overtime, bonus, commission, self-employed, rental, retirement, military, union, nurse, firefighter, and other variable-income cases, because the finalized Income/Employment Processing Checklist task is Addy's calculated income source. Finds the loan itself from loanId, borrowerName, or query, so no separate loan lookup is needed. Returns the task result plus the source documents it used, with links in usedDocuments[].directLink or usedDocuments[].downloadURL. Returns data only for loans the signed-in user is authorized to see, and never returns secrets, raw database content, or other borrowers' or organizations' data.
get_financial_task_answer
Lists the documents already on one loan (uploaded or received), with metadata, status, and categories. Takes a loanId. Documents that are still missing are not included.
get_documents
Reads the loan file and returns an underwriting-grade review grounded in Addy sources: MISMO/1003 loan data, finalized processing checklist results, income, credit, and asset findings, AUS findings, conditions, missing documents, and current guideline overlays. Covers what is wrong or risky in the file, whether it is ready to close, and what to do next.
get_loan_underwriting_review
Returns the prequalification intake data model: the borrower and loan fields to collect, which are required or optional, their allowed values, and the intake conversation policy (one focused follow-up at a time, closely related fields batched, co-borrower questions only when there is a co-borrower, each value marked "confirmed" or "estimated"). Prequalification is information gathering only: Addy makes no approval, denial, pre-approval, qualification, guarantee, or other lending decision.
get_prequalification_fields
Returns the full processing checklist for one loan: every task, complete and open, with its status and result. Takes a loanId. The checklist covers Addy's AI processing tasks; it does not track borrower document uploads.
get_processing_checklist
Lists the Addy organizations the signed-in account belongs to, including the current active or default organization. Useful when the user asks which organization they are in, wants to switch organizations, or works for more than one company.
list_organizations
Lists the document request templates available to the organization (built-in packages plus the organization's custom templates), with the documents each package contains, so the loan officer can browse and choose a package.
list_document_templates
Lists the loan types available in this organization (the same processing-checklist template catalog as Addy's manual "Create" form): built-in system types plus any custom types the organization has added, each with its canonical loanType (checklist template ID), including less common programs (FHA streamline, VA IRRRL, DSCR, bank statement, Canada products, or organization custom types).
list_loan_types
Returns a bounded overview of the loans in the user's pipeline: borrower names, addresses, stages, and page metadata, one page at a time. The widget's Load next page button pages on its own; pass the returned pagination.nextCursor, exactly as returned, only when the user asks for more loans in chat. For large portfolios a broad request first returns a read-only review with the estimated number of accessible loans and suggested refinements, and confirmed: true loads the first page. Small portfolios return the first page right away. organizationId and stage narrow the scope. When the account has several organizations and none is chosen yet, the first call asks which organization to use instead of picking one.
list_pipeline
Lists the documents still missing across the user's loans, by name and status, optionally filtered by borrower name, for questions like "what documents are missing?" or "what does Jane still need to send?". It lists documents only; it does not explain why each one is needed.
list_missing_documents
Lists only the open processing conditions and tasks (not the full checklist), across the user's loans or filtered by borrower name, for questions like "what conditions are still open on Jane's loan?" or "what is outstanding?". No loanId needed. Returns a compact list of open items.
list_pending_tasks
Recommends the right document request package for a loan scenario, so the loan officer does not have to list every document by hand. Takes the loan scenario (loanPurpose, occupancy, employmentStatus, propertyType, loanType) and/or a loanId to fill the scenario from the loan file. Returns ranked templates (built-in packages plus the organization's custom templates) with the documents each one includes and why it matches. The loan officer can add or remove documents before any request is sent.
recommend_document_templates
Sends a document request to a borrower on a loan file by email, text message (SMS), or both. A call without confirmed sends nothing: it returns a review card showing the borrower, recipient, delivery channel (and the mobile number for texts), loan file, documents, and message, and the loan officer sends from that card. When includeUploadLink is omitted, the card also asks whether to include Addy's secure upload link. A call with confirmed: true sends only a request the loan officer was already shown on a review card; otherwise it returns the review card again. Sent requests are tracked on the loan checklist and delivered with the organization's Addy template unless customMessage is given. documentNames can be any document names, including those from the organization's document templates.
request_documents
Finds loans by borrower name, property address, or loan number. Returns matching loans with borrower names, property addresses, loan numbers when available, organization context, links, and the loan ID that the other loan tools take. Searches the active organization when one is selected, otherwise every organization the signed-in user can access, or only the organizationId given.
search
Searches Addy-managed agency, investor, and loan-program guideline overlays: Fannie Mae, Freddie Mac, FHA, VA, USDA, jumbo, DSCR, and investor overlays, covering eligibility, documentation, income, assets, credit, AUS, and underwriting. With a loanId it searches that loan's current program by default; scope "all" searches every program, for broad inventory or comparisons. Results include sourceDocuments with the exact document name and link when available. When sourceDocuments is empty, the result comes from the program guideline context only, not from a specific PDF.
search_guidelines
Sets the active Addy organization for this connection. With organizationName (as the user said it) or organizationId and without confirmed, it returns a review card for switching to that organization and changes nothing. With confirmed: true for the same organization, after the user approved that card, it switches. Later calls use this organization unless an organizationId is passed.
select_organization
Shows a picker of Addy tasks the user can try next: review the loan pipeline, find open conditions or missing documents, review documents already received, review a loan for underwriting, or prepare a borrower document request, with an optional borrower name. The task the user picks is sent from the picker as their next chat message, so this tool reads no loan data and changes nothing. Triggers: "what can Addy do?", "what can I try?", "give me a few tasks to try", "what should I do next in Addy?". When the user already named a borrower, borrowerName pre-fills the picker.
suggest_addy_tasks
Produces a structured prequalification review summary from the collected answers. Each call takes the full set of answers, including corrections, so the summary is replaced cleanly; alternatively the earlier answers go in previousAnswers and only new or changed fields in answers. Each answer is a plain value (treated as confirmed) or an object { value, confidence: "confirmed" | "estimated", note }. Five fields are required to create records: borrower first name, last name, email, loan type, and requested loan amount. The summary separates confirmed and estimated values, lists missing required fields and documents to collect, and says whether the data is ready for record creation. loanType is an exact canonical loan type from the organization's loan types, or a generic agency (conventional, fha, va, usda, jumbo) with a matching loanPurpose; a generic agency without a valid, supported loanPurpose is not ready even when all five required fields are present. An ambiguous or unsupported loan type is reported as loanTypeIssue with readyForRecordCreation=false, separately from the missing-field count. The summary is for the loan officer to review before any Addy records are created, and contains no approval or lending decision.
summarize_prequalification
Switches this connection to a different Addy AI user account. Without confirmed it shows the currently connected account on a review card and changes nothing. With confirmed: true, after the user approved that card, it ends the current Addy session so the user can sign in again as the account they want. For when the user is in the wrong Addy account, wants to log in as someone else, or needs to reconnect.
switch_addy_account
Addy AI ChatGPT Plugin FAQ
How the directory, categories and Discoverability Score work.
Read the methodologyHow do I improve Addy AI's ChatGPT Plugin discoverability?
The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.
What are Addy AI alternatives on ChatGPT?
As of 2026-11-02, Addy AI competes with Ambak Loan Offers, AnyMortgageCalculator, Arthr, AU Home Loan Calculator, Buy Or Rent Simulator, Clara Mortgage Assistant, Danske Bank Housing Calculator, Dr. Klein and 31 more in ChatGPT Mortgages & Home Financing, ranked by public Discoverability Score.
Where is this profile measured?
This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.