M365 Agents Toolkit v6.6.0 got 429 Throttling during teamsApp/extendToM365

Question

The new M365 Agents Toolkit v6.6.0 that just dropped this month. I’m trying to use the teamsApp/extendToM365 action to push my app out, but it keeps spiraling into a 429 Too Many Requests loop. It looks like the toolkit is retrying every 2 seconds without any back-off logic, which is just triggering a permanent throttle from the API. It’s incredibly frustrating because I’m trying to leverage the new v1.25 manifest features, but I can’t even get through the provisioning phase without the deployment failing.

Answer 1

Hi Davies Imelda

As far as I know the teamsApp/extendToM365 step in Microsoft 365 Agents Toolkit can fail with HTTP 429 (Too Many Requests) when the service throttles repeated requests during the “acquire Microsoft 365 title with the app package” operation.

When the client continues retrying quickly, it can keep the caller in a throttled state instead of allowing recovery. Microsoft Graph throttling guidance notes that throttling returns 429 and a suggested delay (for example, via Retry-After), and that immediate retries contribute to ongoing throttling.

Given this, you can try the below approach to see if it can help you:

1. Stop the teamsApp/extendToM365 run as soon as 429 starts, then retry only after the service’s wait time (for example, Retry-After). Immediate retries prolong throttling.
2. If you need to keep working now, remove/comment out teamsApp/extendToM365 in the lifecycle YAML so provisioning can complete without hitting the throttled “extend” endpoint.
3. Update (or temporarily switch) the Agents Toolkit version using “Install Specific Version…” in VS Code to pick a newer build (or roll back) if the issue persists in your current version.

Hope my answer will help you.

---

If the answer is helpful, please click "Accept Answer" and kindly upvote it. If you have extra questions about this answer, please click "Comment".   

Note: Please follow the steps in our documentation to enable e-mail notifications if you want to receive the related email notification for this thread.  

Answer 2

@Davies Imelda In v6.6.0 of Agents Toolkit, the teamsApp/extendToM365 action should be polling app upload status every 7s instead of 2s before to workaround the 429 error. Is it still 2s on your side and from what behavior you can tell it's 2s?

Answer 3

The behavior matches generic service-side throttling: the toolkit is sending frequent calls, the backend responds with HTTP 429, and the rapid retries keep the client in a throttled state.

From the available guidance, the only supported mitigation is to slow down calls and honor the service’s throttling signals until the throttle window expires:

1. Stop the current teamsApp/extendToM365 run and wait
   • Microsoft Graph and other Microsoft 365 services respond with HTTP 429 and a suggested wait time in the response headers.
   • When throttling occurs, further requests from the same client are limited for a period of time. Continuing to call every couple of seconds keeps the client in a permanently throttled state.
   • Wait at least as long as the service’s throttle window (typically several minutes; in some cases longer) before retrying the operation.
2. Retry after the throttle window
   • After waiting, run the teamsApp/extendToM365 action again.
   • If the toolkit version continues to retry too aggressively and re-enters throttling, repeat the wait‑and‑retry cycle rather than repeatedly triggering the action.
3. General throttling guidance that applies to the toolkit scenario
   • Microsoft Graph and other APIs throttle when request thresholds are exceeded and return HTTP 429 with a recommended delay. Clients are expected to:
     • Back off when 429 is received.
     • Respect the suggested wait time before sending new requests.
   • High-volume or bursty operations (such as provisioning and deployment) are more likely to be throttled, especially when many write operations are involved.
4. If repeated throttling persists
   • Run the provisioning/deployment less frequently (avoid rapid successive runs while testing changes).
   • Consider separating steps (for example, avoid combining many provisioning changes into a single rapid sequence of runs) so that each run performs fewer operations.
   • If the issue continues even with long pauses between runs, raise it as a product issue for the Microsoft 365 Agents Toolkit so that the retry/back-off behavior can be adjusted in the tool itself.

These steps do not change the toolkit’s internal retry logic, but they align usage with the documented throttling behavior of the underlying services and avoid a permanent 429 loop while using teamsApp/extendToM365 to deploy a v1.25 manifest–based app.

---

References:

• Microsoft 365 Agents Toolkit
• Microsoft Graph throttling guidance