Skip to main content

​
Overview

The Aspire integration gives your agents read-only access to your Aspire field-service-management tenant. Agents can query Properties, Opportunities, Work Tickets, Invoices, Contacts, Companies, Routes, Services, and the rest of Aspire’s data via OData filters — to answer operational questions, reconcile records, and feed analysis, without ever modifying anything in Aspire. The integration is read-only by design for the initial release: the underlying client only exposes GET, so agents cannot create, update, or delete Aspire records.
This is Aspire by youraspire.com (landscape & field service management) — not Microsoft .NET Aspire and not the Aspire fintech product.

​
Prerequisites

  • An Aspire account with API access
  • An Aspire Client ID and Secret, generated under Account Settings > API Keys in Aspire
  • An Aster admin to connect the integration (the connection is shared org-wide)

​
Setup Guide

1

Generate API credentials in Aspire

In Aspire, go to Account Settings > API Keys and create a key. Copy the Client ID and Secret.
2

Connect Aspire in Aster Agents

Navigate to Control Hub > Integrations, locate the Aspire card, click Connect, and paste the Client ID and Secret. Aster validates them against Aspire (token exchange + a tenant lookup) and echoes back your tenant’s legal name on success. Aster handles token exchange and refresh from there.
3

Add Tools to Your Agent

Edit your agent and enable Aspire Get Schema and Aspire Query. The agent discovers the available collections and fields at runtime, then queries them.

​
Available Tools

  • Aspire Get Schema (aspire_get_schema) — Introspects Aspire’s live OpenAPI spec to discover the available resource collections and their fields. Agents call this first to look up collection and field names instead of hard-coding them.
  • Aspire Query (aspire_query) — OData-style read against any whitelisted Aspire collection (Properties, Opportunities, WorkTickets, Invoices, Contacts, Companies, Jobs, Services, …). Supports $select, $filter, $expand, $orderby, and $top/$skip paging. Auto-paginates up to a configurable row cap; large result sets spill to a CSV attachment while the inline payload is truncated to a token budget, so big pulls don’t balloon the agent’s context.
Read-only. This release exposes reads only — agents cannot create, update, or delete Aspire records. The HTTP client only permits GET; write capability is intentionally not wired up.

​
Example Prompts

Once connected, agents can answer questions like:
  • “List active Opportunities over $5,000, newest first.”
  • “How many open Work Tickets do we have by Route this week?”
  • “Pull every Property for the Smith account with its service plan.”
  • “Export all Invoices from last quarter for analysis.” (the full set spills to a CSV the agent can process in Python)

​
Notes & Limitations

  • Read-only — no writes in this release, by design.
  • Page size cap — Aspire hard-caps $top at 1000 rows server-side; the tool auto-paginates to reach larger totals.
  • No webhooks — Aspire has no push/webhook mechanism, so agents poll. For scheduled monitoring, match your task cadence to how fresh you need the data.
  • Single environment — Aspire has no sandbox; credentials point at your live tenant. Because the integration is read-only, that’s safe by construction.

​
Security Considerations

  • Org-level connection — one admin connects once for the whole organization. The Secret and tokens are stored server-side and redacted when the connection is read back in the UI.
  • Read-only by construction — the client cannot mutate your Aspire tenant.
  • Grant the Aspire tools only to agents that need them.
  • Rotate credentials periodically from Aspire’s API Keys screen and reconnect.

​
Troubleshooting

  • Double-check the Client ID and Secret were copied exactly from Account Settings > API Keys in Aspire.
  • Confirm API access is enabled on your Aspire account.
  • Run Aspire Get Schema to confirm the exact collection name and field names — they’re case-sensitive OData identifiers.
  • Loosen your $filter; an over-narrow expression can legitimately match nothing.
  • Aspire pages at 1000 rows, so big collections take several sequential calls. The full set is written to a CSV attachment; the inline preview is intentionally truncated to keep the agent’s context small. Have the agent process the CSV (e.g. in execute_python) rather than reading every row inline.
  • Execute Python — process large Aspire result sets (the CSV spillover attachment) with pandas.