Setting Up The Wazzi Data Connector

First-party read-only access to your organization's synced PMS data — reservations, properties, owners, work orders, amenities, revenue KPIs, and LOM contacts — with no credentials to configure. Always available once your data sources have synced.

August 12, 2026
7 min read

Path: Connectors → Wazzi Data → Configuration

What this connector does

Wazzi Data is a first-party, read-only MCP connector that exposes your organization's synced PMS data — reservations, properties (units), owners, work orders, amenities, revenue KPIs, and local operations managers (LOMs) — directly to AI assistants. It's the same data that powers the dashboard and drives internal agents; the connector just makes it queryable via natural language.

Because Wazzi Data reads from Wazzi's own database, there's nothing to configure and no external credentials to enter. Every request is org-scoped server-side, so cross-organization reads are impossible regardless of which client calls the connector. Turn it on for a user's group and they're ready to query.

The connector exposes 10 tools grouped by domain:

  • Reservations (3 tools) — get one/many by id, get all reservations on a unit (optionally bounded to a year), and read PMS-side reservation flags (VIP, Owner Stay, DND, etc.).
  • Properties (units) (2 tools) — get one/many by unit_id (with property custom fields merged in), or list every unit for the org with pagination + filters.
  • Work orders (1 tool) — maintenance tickets filterable by unit, status, critical flag, and date range.
  • Amenities & revenue (2 tools) — per-unit amenity attachments, plus monthly revenue KPIs (GBV, ADR, occupancy, booked nights) sourced from the KeyData sync.
  • Owners & LOMs (2 tools) — property owner contact info (by owner_id or resolved through a unit) and Local Operations Manager assignments including market, backup, and GM contacts.

Before you start

Because Wazzi Data reads from Wazzi's synced PMS tables, this connector is only useful if you have at least one data source configured and synced — otherwise every tool call returns an empty result. The relevant sources:

Once a data source is configured and its initial sync completes (usually a few minutes to an hour depending on portfolio size), Wazzi Data becomes queryable.

What this connector unlocks for your team

Once configured, members with the right MCP permissions can ask Claude / ChatGPT / Cursor things like:

  • "Show me every reservation checking in tomorrow across the Phoenix market."
  • "Pull the full reservation history for unit 1042 for this year."
  • "Which units have open critical work orders that haven't been closed in more than 3 days?"
  • "What was our T12 gross booking value across all units in the Scottsdale portfolio?"
  • "List every unit with a private pool and a hardline gas BBQ."
  • "Who's the LOM for the Camelback market and what's their phone number?"
  • "Get me the owner's email for unit 1042 so I can loop them in on this issue."

The exact set of available actions depends on which MCP toggles are flipped on for the user's group in Wazzi — see Managing Permissions.

Steps — enabling Wazzi Data in Wazzi

1. From the Connectors catalog, click Configure on the Wazzi Data tile.

You'll land on Wazzi Data's Configuration tab:

Wazzi Data connector Configuration tab

2. Read the notice — there are no credentials to enter.

The configuration tab is intentionally empty aside from an information notice: this connector reads from Wazzi's synced PMS tables and doesn't require any external API key.

3. Click Test Connection.

Wazzi runs a quick count query against your synced units and reservations. On success you'll see:

  • "Connected. 1,247 units, 8,932 reservations available." — data is synced and ready.
  • "Connected. No PMS data has been synced yet — enable Streamline (or another PMS) under Settings → Data Sources…" — the connector is wired up, but there's nothing to query yet.

4. Click Save Configuration.

Wazzi Data moves to the Active section of the catalog and the dot turns green.

Troubleshooting

  • Test Connection returns "No PMS data has been synced yet". The connector is wired up correctly; your organization just doesn't have any PMS data yet. Configure Streamline (or another PMS) under Settings → Data Sources and let the initial sync complete.
  • Tool calls return empty arrays. Either the filter is too restrictive (widen the date range or remove optional filters), or the specific data category hasn't synced yet. Work orders, amenities, and revenue each have their own sync — check Settings → Data Sources to confirm the category you're querying is enabled.
  • "This connector requires an authenticated session." The tool was called without a valid Wazzi org context. Reconnect the MCP from your client — it may have lost the auth binding.
  • Revenue values look stale. KeyData's monthly revenue sync runs on a schedule. If you need current-month data, wait until the next sync cycle or trigger a manual refresh from Settings → Data Sources → Revenue.
  • Owner information is missing on a unit. The owner is either not linked in Streamline, or the owners sync hasn't run yet. Check Settings → Data Sources and confirm the owners category is enabled.

Best practices

  • Combine Wazzi Data with the source PMS connector. Wazzi Data is fast (it reads from a synced local table) but is only as fresh as the last sync. For live data (current availability, unposted transactions, changes made in the last few minutes), pair it with Streamline / Track / etc. for the live path.
  • Use list_properties with filters before drilling in. The full inventory can be hundreds or thousands of rows — filter by city, state, location_area, or property_group first, then use the returned unit_ids to fetch reservations / amenities / revenue.
  • Set include_custom_fields: false on large portfolios. The default is true; on orgs with many custom fields, the response can be substantial. Turn it off when you only need the core unit fields.
  • Use year-scoped reservation queries. get_reservations_by_unit defaults to the current year — pass year for historical windows or include_all_years: true only for full audit pulls.

Frequently asked questions

Do I need to configure anything?
No — that's the point. Wazzi Data reads from your own synced tables, so there are no external credentials to enter. You just need the underlying data sources (Streamline / KeyData / etc.) configured and synced.

Can users from another organization see my data?
No. Every tool call filters by organization_id at the SQL level, using the authenticated org from the calling token. Cross-org reads are impossible regardless of which client or token reaches the server.

Why can't I write data through this connector?
Wazzi Data is deliberately read-only. Writes need to go through the source PMS connector so they land in the system of record (Streamline / Track / etc.) — writing back to Wazzi's local mirror would create split-brain drift the next sync would silently clobber.

How fresh is the data?
Data sources run on their own sync schedule — typically every 15 minutes to an hour depending on the category. Reservations are the freshest (near-realtime for most PMSs); revenue KPIs update monthly.

Which fields are exposed on properties?
get_properties / list_properties return core unit metadata (name, code, address, geo, bed/bath/max_occupants, amenities booleans, location hierarchy) plus every property custom field configured for your org, merged into a custom_fields object. Operational / PII fields (lock codes, WiFi passwords, owner contact, commission data) are NOT included — those are guarded and only visible to the internal dashboard / agents.

What's next

Was this article helpful?