MCP App Store
by KeenEthics
Presuo icon

Presuo MCP Server for ChatGPT

By RecreaVision

Overview

Manage the lifecycle of construction budgets, from creation to field progress reporting in ChatGPT. Shape a construction budget proposal with the help of ChatGPT. Ask natural-language questions about existing budgets, work packages, quantities, measurements, costs, and project activity. Quickly find items, inspect quantity breakdowns, understand remaining work, and navigate large budgets without digging through menus. Team members can report draft executed quantities and collaborate through comments and approvals, keeping project progress current from the office or jobsite. Get your Presuo account, create new proposals or upload existing budgets, and access them securely in ChatGPT.

How to connect

  1. ChatGPT: Open the Presuo listing in the ChatGPT app directory and choose Connect. ChatGPT listing Presuo exposes 20 tools here.

Tools

activity.list

ChatGPT
Comments, approvals and system events (certifications included) of a line item or budget, newest first. For what is pending, use approvals.list. Budget reads cover 30 days unless since/until are set (90 max). Paged: if has_more, say so and use next_cursor only as needed; an empty page with has_more is not the end. If text_truncated, the full comment is in Presuo. Measurement details show current values. Read-only.

approvals.close

ChatGPT
Marks a pending approval as reached for one confirmed line item. Use only after explicit user confirmation immediately before calling this tool. Send a new idempotency_key for each confirmed closure and reuse it only to retry that same call. If no approval is pending, nothing changes and the receipt reports changed=false. A comment alone does not resolve an approval.

approvals.list

ChatGPT
Shows current approval status and approval history for one line item or a whole budget. Use to find pending approvals or review who requested or resolved them. Budget counts cover every line item; item lists and history are paged. Approval status is separate from measurement certification. Supports time filters and pagination. Read-only; does not change the activity state in Presuo.

approvals.request

ChatGPT
Requests approval for one confirmed line item. Use only after explicit user confirmation immediately before calling this tool. Send a new idempotency_key for each confirmed request and reuse it only to retry that same call. If approval is already pending, returns the current state without creating a duplicate request.

budget.browse

ChatGPT
Lists the contents of one chosen budget in budget order, one page at a time: its sections by default, one section's line items with category_id, or every line item with all_items=true. Use it when the user asks what a budget or section contains; use budget.find for a named or described item. A page is a view, not the whole budget: report count_total and has_more, and fetch further pages only as the question needs. For measurements, use measurement.list_for_line_item. inclusion: rows not included or included elsewhere, 0 in amounts. Show display_reference or names to users, never IDs. Read-only.

budget.editor.open

ChatGPT
Opens the existing fullscreen Presuo editor for one accessible, editable committed budget. For a clear user request naming an explicit budget ID, call directly; the ID is only a locator and Presuo checks access, edit permission, lifecycle, and budget type. For names or conversational references, first use budget.list_accessible, and ask the user to choose if multiple matches remain. Do not call for read-only questions or unclear intent. This does not edit through conversation and cannot open a ChatGPTBudget. The host UI consumes handoff_url as a one-time iframe bootstrap; never present it as a user-facing link. Use direct_url for a clickable fallback; it opens the ordinary Presuo editor and may require the user's Presuo login.

budget.find

ChatGPT
Finds sections, line items or measurements that the user names or describes inside one chosen budget: exact codes (leading zeros in numeric segments ignored), exact labels, then partial text. Use it for any item or section reference; locate the budget itself with budget.list_accessible. outcome=resolved is one exact match and final for reading; ambiguous lists exact candidates; matches are partial hits, never a write target until the user chooses; none is an exhaustive no-match. When has_more is true, say the list is partial and ask the user to narrow it. Before any write, read the chosen line item with measurement.list_for_line_item. Non-null inclusion: not included or included elsewhere (inclusion.set), 0 in amounts. Show display_reference to users, never IDs. Read-only.

budget.list_accessible

ChatGPT
Lists ezbudget budgets visible to the authenticated user. Use when the user asks to list, show, browse, select, or navigate their budgets or projects. Returns compact budget names, project names, statuses, and pagination metadata. This tool is read-only and does not return budget line items, measurements, costs, or financial details. IDs are internal for tool chaining and should not be displayed unless the user explicitly asks for them.

category.list_for_budget

ChatGPT
Lists categories within a selected ezbudget budget visible to the authenticated user. Use when the user asks to browse, show, or select categories inside a budget. Read-only. When answering the user, describe results as Presuo information. Do not mention API responses, tool calls, payloads, or schemas. IDs are internal for tool chaining and should not be displayed unless the user explicitly asks for them.

comments.create

ChatGPT
Posts a plain-text comment on one confirmed line item. Use only after showing the final comment and target and receiving explicit user confirmation immediately before calling this tool. Send a new idempotency_key for each confirmed comment and reuse it only to retry that same call. The comment becomes part of the line item's activity history. A comment does not itself request or resolve an approval.

draftbudget.discard

ChatGPT
Permanently discards the active private draft when the user explicitly requests it. This does not delete a committed budget.

draftbudget.get

ChatGPT
Reads the active saved proposal. Set include_promotion_review when preparing to create a budget; this call changes nothing.

draftbudget.promote

ChatGPT
Creates an In preparation budget only after the user explicitly approves the exact promotion review returned by draftbudget.get.

draftbudget.save

ChatGPT
Creates or saves the complete proposal snapshot. Null draft ID and revision create a draft; existing ID and current revision update it. Omitted rows are deleted, so preserve every untouched row and ID. A measurement's optional inclusion marks it not_included (not offered) or included_elsewhere in target_id, a saved row; totals ignore both, which are neither cancelled nor out of scope. Links whose target is deleted or excluded become not_included: report inclusion_conversions to the user.

inclusion.set

ChatGPT
Sets the inclusion state of a measurement, or of every eligible measurement of a line item or category: included (charged here, the default), not_included (not offered by this budget) or included_elsewhere (done, paid through target_kind/target_id, an included row of the same budget). Totals ignore rows not included or included elsewhere; their price is kept. These states are not cancellation and not out-of-scope work: never call them cancelled or out of scope. Preparation and validation allow every state; execution only links or unlinks pending rows (no not_included); completed budgets are read-only. included clears target and note. Rows included in a row that stops being included are converted automatically and listed in converted_dependents: tell the user. Call only after the user confirmed.

line_item.list_for_budget

ChatGPT
Lists line items within a selected visible ezbudget budget. Use when the user asks to browse all line items in a budget. Read-only. When answering the user, describe results as Presuo information. Do not mention API responses, tool calls, payloads, or schemas.

line_item.list_for_category

ChatGPT
Lists line items within a selected category of a visible ezbudget budget. Use when the user asks to browse, show, or select line items in a category. Read-only. When answering the user, describe results as Presuo information. Do not mention API responses, tool calls, payloads, or schemas.

line_item.resolve

ChatGPT
Resolves a user-provided line-item reference within visible ezbudget budget scope. Use when the user refers to a line item by code, title, or shorthand and target disambiguation is needed before continuing. Read-only. When answering the user, describe results as Presuo information. Do not mention API responses, tool calls, payloads, or schemas. IDs are internal for tool chaining and should not be displayed unless the user explicitly asks for them.

line_item.search_in_budget

ChatGPT
Searches categories, line items, and measurements within a selected visible ezbudget budget using existing budget search behavior. Read-only. When answering the user, describe results as Presuo information. Do not mention API responses, tool calls, payloads, or schemas.

measurement.clear

ChatGPT
Clears reported values or the note of one measurement. Clearing anything but the note needs a matching measurement.preview first. Call only after the user confirmed exactly what is cleared.

measurement.list_for_line_item

ChatGPT
Lists measurements attached to a selected ezbudget line item visible to the authenticated user. Use when the user asks to browse, show, or select measurements for a line item. Read-only. When answering the user, describe results as Presuo information. Do not mention API responses, tool calls, payloads, or schemas.

measurement.preview

ChatGPT
Previews a measurement.report or measurement.clear sent with the same arguments (no idempotency_key). Required before a report with replace, copy_planned_quantity, all_measurements or over 25 measurements, and before clearing more than the note. Show it, get explicit confirmation, then write the same arguments within 10 minutes. If full_list is false, state the counts, totals and overwrite_count and say the samples are partial. Read-only.

measurement.report

ChatGPT
Reports progress or a note on one line item's measurements (listed with revisions, or all_measurements). explicit_quantity sends values; copy_planned_quantity copies planned quantities; a note-only append sends only payload.note. Call only after the user confirmed; some writes need measurement.preview first. Does not alter pricing or finalize the line item.

measurement.revert

ChatGPT
Restores the reported values and note one measurement had before its last report or clear. Call only after the user confirmed the rollback.

project.list_accessible

ChatGPT
Lists projects eligible as a budget destination.

session.start

ChatGPT
Start a short-lived Presuo MCP browsing session for the authenticated user. Call this first, then pass the returned session_id to subsequent Presuo tools (for example budget/category/line-item/measurement lookups) so requests stay scoped to the same conversation context until the session expires.

Capabilities

WritesInteractive

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/presuo"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/presuo)