Category: Forge Platform and Tools
We’re a Marketplace partner building a Forge app for enterprise Confluence/Jira customers in Korea.
With usage-based pricing for Rovo credits and Automation steps starting Dec 3, 2026, our customers are asking us to surface usage and cost controls inside our app so admins don’t have to go to Atlassian Administration for every check.
To be clear up front: this is about the *customer’s* meters (Admin → Insights → Platform usage), not the Forge platform charges we pay as a vendor.
We know about the Developer Console usage/commerce APIs and the August thread on Forge cost alerts (FRGE-2252 / FRGE-2253) - those cover our own costs and don’t help here.
What we’ve checked so far:
- Support docs (Platform usage, How Rovo credits work, How is your automation usage calculated): customer usage is only visible in Admin → Insights → Platform usage, plus a manual CSV export (up to 3 months / 1M events). “Allow extra usage” and spending limits are admin-UI only.
- Cloud Admin REST APIs (Organizations, User management, User provisioning, API access) and the Cloud Admin changelog through 2026: nothing related to usage, meters, allowances, or limits.
- Forge product REST API list: nothing exposing meter usage either.
- Automation Rule Management API: GA, but admin-token only; Forge/OAuth 2.0 apps can’t call it (per the April reply in the GA announcement thread).
So as far as we can tell, there is no supported way for a Forge app to read a customer org’s current meter usage, allowance, or limit settings, or to react when a meter is paused.
Questions:
1. Is that understanding correct — no API (REST or Forge) today to read an org’s meter usage (Rovo credits, Automation steps) or its allowance/limit settings?
2. Is exposing this to Forge apps on the roadmap? Ideally as a Forge scope (org-scoped read of Platform usage, or at least the usage attributed to the calling app) rather than via an org API key. Enterprise customers won’t hand an org-wide admin API key to a Marketplace app, so an admin-key-only endpoint wouldn’t be usable for us. Is there an RFC we should follow?
3. For usage our own app generates (Rovo agent / Teamwork Graph calls, automation rules our app triggers), is there any way for the app to see how much it actually consumed, so we can meter and gate it accurately instead of estimating from the documented credit/step rules?
4. Are there any events or webhooks when an org hits 80% / 100% of a meter, or when a meter is paused because extra usage is disabled? Today the only signal we have is failed calls.
5. (Related) Any update on Forge / OAuth 2.0 access to the Automation Rule Management API? Even a scoped version limited to rules the app itself created would cover our case — customers want to enable/disable app-created rules from inside the app for cost control.
For now we plan to meter our own app’s calls ourselves and enforce per-space / per-user caps in the app, with a deep link to Platform usage for org-level controls. If there is a better-supported pattern, we’d rather follow it — and given the Dec 3 date, we’d appreciate knowing even a rough direction so we don’t build against the wrong assumptions.
References we’ve read:
- What is Platform usage? | Atlassian Support
- How Rovo credits work | Rovo | Atlassian Support
- How is your automation usage calculated | Cloud automation Cloud | Atlassian Support
- https://developer.atlassian.com/cloud/admin/changelog/
- Forge Platform Usage Costs: Alerts & Limits
- Automation Rule Management API is now Generally Av... - Atlassian Community
