Spectora API & MCP Early Access FAQs

The Public API and AI Connector are currently in early access and available to paid Spectora customers.

To join the early access group, request access via the waitlist.

It's free during early access — no additional cost while we're in this phase.

We're still working through what pricing will look like when the feature is fully released, and we'll share those details well in advance so you have plenty of time to plan.

Pricing isn't finalized yet. We expect there to be a free tier and possibly several paid tiers when we move out of early access.

You can have API access for multiple companies.

Each Spectora company instance has its own login, API keys, and webhook subscriptions.

When tiered pricing is in place, pricing may vary between companies depending on differences in plan or usage.

We recommend creating a dedicated staff account with full permissions specifically for API use. Then:

From there, you can test endpoints under the API Reference tab or configure webhooks under Webhook Configuration.

The API key inherits the permissions of the user who created it.

If that user is later deactivated, the key stops working.

A dedicated API user keeps your integrations stable and easy to manage over time.

Not at this time, outside of the limitations imposed by the permissions of the user who created the key.

Yes! We have walkthroughs covering:

No. This is a known request and is under active internal discussion.

Full report content — narratives, comments, and photos — isn't exposed via the API today.

You can get inspection metadata, and you can build the client portal link using the slug field included in the inspection response:

Adding the report link directly to the response is something we are working on.

Not yet. Calendar availability endpoints aren't exposed today — this is one of the most requested additions and is being assessed for prioritization.

Not yet. A services endpoint is in our planning but not yet available.

The current inspection response does include service and add-on names in comma-delimited format, so you can see what was on an inspection — but not the per-service prices.

Yes — most agent fields are available via the API, including agency name, buying and selling inspection counts, and tags.

To get agent details inside an inspection response, use the include parameter when calling the inspection endpoint.

Yes — use the include parameter when retrieving an inspection to get the buyer's email, phone, and address. Without it, you'll only see the buyer's name.

The primary inspector name and ID is available via the API.

A dedicated Inspectors endpoint is on our roadmap.

Not today — the API exposes a paid boolean only.

Exposing granular payment data is on our list.

Not today.

We're evaluating exposing agreement PDFs, template names, and signed status by party.

Two things:

.jpg, .jpeg, .png, .gif, .pdf, .zip, .txt, .mp3, Word, Excel, and PowerPoint files.

Yes — uploads must be sent as multipart/form-data with key/value pairs in the format data[attributes][fieldname].

The developer portal supports this when you switch the Body to multipart. The walkthrough video shows the full flow.

The primary use case is third-party vendor integrations — for example, lab partners automatically posting completed reports back into Spectora where they appear in the client portal.

You can also use it for your own workflows, though it's most powerful for automated flows rather than one-off uploads.

Payloads are intentionally minimal — they include the event type, the object ID, and the company ID. To get full details, make a follow-up API call using the object ID. This is by design for speed and reliability.

Yes.

No. Only your outbound API calls count against your quota — incoming webhook deliveries do not.

The AI Connector uses the Model Context Protocol (MCP) — an open standard that lets AI assistants like Claude and ChatGPT connect directly to your Spectora data.

Instead of writing code to pull data, you connect your Spectora account to an AI assistant and ask questions in plain language. No programming required.

Generate an API key from the developer portal at developer.spectora.com, then follow the setup instructions on the MCP Server tab. Setup typically takes about 5 minutes.

Claude Desktop

ChatGPT

Yes. Write endpoints are intentionally disabled through the AI Connector — this prevents AI tools from creating or modifying records in your Spectora account without you doing it deliberately.

Yes — both the Public API and the AI Connector are bundled in the current early access offering.

Yes. Each company needs its own API key and its own entry in your AI tool's configuration. Name them clearly — for example, spectora_company1 and spectora_residential — so you can reference the right one. Here's how:

A few high-value things inspectors are using it for today:

Yes — there is both a per-minute rate cap and a monthly call volume quota.

Use delta queries — only pull what's changed since your last sync rather than pulling full history each time.

Combining webhook events with targeted API calls by ID is typically much more efficient than periodically listing all records.

A 429 Too Many Requests response means you've hit your rate limit.

Build retry logic with exponential backoff into your integration — when a 429 comes back, wait a moment and try again, waiting a little longer each time.

If you believe your quota needs to be increased, reach out to us on the chat bubble or at support@spectora.com.

All API requests use Bearer token authentication. Include your API key as a Bearer token in the Authorization header of every request.

Delete the compromised key first, then generate a new one and update your integration:

Yes — all API calls go over HTTPS.

No. We do not use customer data to train any AI models.

developer.spectora.com

Yes — the developer portal has an interactive playground under the API Reference tab. It uses your real Spectora data, so be thoughtful about what you test.

Multipart endpoints like attachment upload require switching the Body type to multipart in the playground.

Yes — under Webhook Configuration. The video walkthrough shows the full flow.

We love hearing from you. Reach out via the in-app support bubble or email josh@spectora.com directly.

For access or API key issues, reach out via the in-app support bubble or email support@spectora.com

For technical issues, include the request URL, the parameters you sent, and the response you received — but never include your API key in any support request