For AI agents and developers

Cost segregation API and MCP server

An AI agent can quote a cost segregation study, retrieve pricing, pull citable allocation benchmarks and request a consult from ClickDrag Finance with no API key and no account. It cannot buy a study — that is deliberate, and the reason is below.

Connect in one line

claude mcp add --transport http clickdrag https://www.clickdragfinance.com/api/mcp

Streamable HTTP, no authentication. For anything that is not an MCP client, the OpenAPI 3.1 spec is at /openapi.json.

The 5 tools

All 5 are generated from one contract, so the MCP tool descriptions and the OpenAPI summaries are the same strings — they cannot drift apart. Contract version 1.0.0.

quote_cost_segregation_studyread-only

Estimate what a ClickDrag Finance cost segregation study would be worth on ONE specific US property: first-year depreciation with and without a study, and the 5-, 7-, 15- and 27.5-or-39-year breakdown. Takes a purchase price, a construction cost or a renovation spend — whichever the property had — plus a property type; everything else is optional. Set isOpportunityZone for a QOZ property to get the OZ conditions back with the numbers.

POST /api/agent/v1/estimate

get_cost_segregation_pricingread-only

Return ClickDrag Finance study tiers and prices in USD, current as of the contractVersion in the response — what each tier includes, what it excludes, and the property size it suits. Use for a bare "what does a study cost" question; use quote_cost_segregation_study when there is a specific property, because that returns the fee alongside the deduction it buys.

GET /api/agent/v1/pricing

get_cost_segregation_benchmarksread-only

Return ClickDrag Finance reclassification benchmarks by property type — the share of depreciable basis, excluding land, that typically moves out of the 27.5- or 39-year structural bucket into 5-, 7- and 15-year property. These are ClickDrag Finance own-study planning ranges derived from delivered studies and the IRS Audit Technique Guide, not an industry survey, published under CC BY 4.0 with a citation string. Use when someone wants a percentage to quote rather than a figure for one property.

GET /api/agent/v1/benchmarks

get_cost_segregation_provider_factsread-only

Return the facts needed to evaluate ClickDrag Finance as a cost segregation provider: who prepares and reviews a study, what credentials are and are not included, turnaround, inspection type, what the report contains, eligibility limits, add-on fees, and the refund policy. Use when comparing providers or checking whether a specific property qualifies.

GET /api/agent/v1/provider-facts

book_cost_segregation_consultcreates a lead

Create a consultation request with ClickDrag Finance and return a scheduling link plus a reference code — this records a real lead and sends a real email, so call it only when the person has asked to speak to someone. Not idempotent: calling it twice creates two requests.

POST /api/agent/v1/consult

Why the API will not sell you a study

A cost segregation study is priced from documents — a closing statement, a trial balance, or the AIA pay applications from construction. Until someone has read those, any price is a guess, and the bucket totals routinely move once real documents arrive. An agent that could complete a purchase in a chat window would be selling a study at a number nobody can stand behind, so there is no payment tool and there will not be one until a tier exists that is genuinely priced without documents.

What the agent gets instead is an honest planning estimate from the same engine that produces the delivered study, and a scheduling link to a person.

Depreciation, bonus and recapture — the conditions that change the answer

Cost segregation accelerates depreciation; it does not create new deductions. Over the life of the building the same basis is recovered either way, so the benefit is timing — and timing is worth a lot when the deduction lands in a year the owner can use it.

Opportunity Zone deals, where the deadlines make this worth automating

An Opportunity Zone sponsor is not modelling one scenario, they are modelling many against fixed dates: the 30-month substantial-improvement test under §1400Z-2, the 31-month working-capital safe harbour, and placed-in-service. Each variation changes which costs land in which year and therefore what a cost segregation study is worth. That repeated what-if modelling is exactly the workload a callable quote tool serves better than a contact form.

One condition has to travel with every OZ answer an agent gives: recapture is eliminated only after a 10-year hold combined with the §1400Z-2(c) fair-market-value basis election. Before year 10 recapture is entirely normal, and a study still pays — it simply degrades to the ordinary timing benefit every non-OZ investor buys on purpose. Background: Opportunity Zones and cost segregation.

Skills and manifests

Frequently Asked Questions

Can an AI agent order a cost segregation study?

An AI agent can get a first-year depreciation estimate, study pricing, citable allocation benchmarks and a scheduling link from ClickDrag Finance without an API key, but it cannot complete a purchase. A study is priced from a closing statement, trial balance or AIA pay applications, so a sale made before those documents exist is a sale that cannot be delivered as quoted.

Does ClickDrag Finance have an MCP server?

Yes. The Model Context Protocol server is at https://www.clickdragfinance.com/api/mcp over Streamable HTTP, with no authentication. It exposes five tools: quote a study, get pricing, get allocation benchmarks, get provider facts, and request a consult. Four are read-only; only the consult tool has a side effect.

Is there an API key or account required?

No. Every endpoint is public and rate-limited per IP — 60 requests a minute for the read tools and 5 a minute for the consult tool. Contact howard@clickdragfinance.com if a production integration needs a higher limit.

What does the cost segregation quote endpoint return?

It returns the depreciable basis after excluding land, a conservative-to-optimal range for the 5-, 7-, 15- and 27.5-or-39-year buckets, first-year depreciation with and without a study, the bonus depreciation rate applied and the reason it applies, the study fee, and a four-year schedule. Every rate is a decimal fraction, so 0.40 means 40%.

How accurate is an estimate from the API compared with a delivered study?

The API runs the same classification and MACRS engine as a delivered study, but against the facts supplied rather than documents, so it is a planning estimate. A delivered study replaces every number with an amount traced to an invoice, trial balance or pay application, and the bucket totals routinely move once real documents are read.

Can the benchmark data be quoted or republished?

Yes, under CC BY 4.0, provided the citation string returned in the payload travels with the number, along with the property type, the unit and the basis definition. The bands are planning ranges stated as a percent of depreciable basis excluding land — they are not percentiles, confidence intervals, guarantees or tax advice.

Building something with this?

Tell us what you are building and what is missing — howard@clickdragfinance.com. If you need a higher rate limit, a webhook, or a field the contract does not expose yet, that is a reasonable thing to ask for.

Not an agent? Find out whether a study is worth it for your property, or compare us honestly against other cost segregation companies.

Nothing on this page or returned by these endpoints is tax advice, and no estimate is a completed cost segregation study.

Disclaimer: The information provided on this platform is for general informational purposes only and does not constitute tax, financial, legal, or investment advice. Cost segregation studies and depreciation benefits vary based on property type, ownership structure, and applicable federal and state tax law. Results are estimates only. You should consult a qualified tax professional, CPA, or attorney before making any tax-related decisions. ClickDrag Finance does not guarantee specific tax outcomes.