azure ad AADSTS70011: The provided request must include a scope input

Not specified by the original author.

TL;DR

Add the scope parameter to the failing request. Microsoft's v2 endpoint requires scope on authorize and token calls, and leaving it out fails with AADSTS70011. Include the resource scopes you need (or the default scope) and the error clears.

azure ad AADSTS70011: The provided request must include a scope input parameter

Steps

1. Identify which call fails: the authorization redirect or the token exchange. Expected: you know which request lacks the parameter.
2. Add the scope parameter to that request. For Microsoft Graph access use the Graph scopes you need; for your own API use its scopes, or append the default scope suffix to get all statically consented permissions. Expected: the request now carries an explicit scope.
3. Keep the scope consistent between the authorize request and the token request. Expected: both calls send the same scope value.
4. Retry the flow. Expected: AADSTS70011 is gone; if a new error names a specific scope, that scope needs admin consent or does not exist.
5. If you migrated from the v1 endpoint, audit every token call: v1 used a resource parameter where v2 requires scope. Expected: no remaining v1-style calls.

Use this when

• The error code is exactly AADSTS70011.
• A generated client was built against v1-style docs (resource parameter) and pointed at the v2 endpoint.
• The token request was hand-assembled and scope was simply omitted.

Not for this skill when

• The error names a specific scope as invalid - that is a scope-content problem, not a missing parameter.
• Consent or admin approval is the blocker - those are different error codes.
• The provider is not Microsoft identity platform; scope requirements differ elsewhere.

Variant phrasings

• AADSTS70011 missing scope parameter
• Azure AD token request must include scope
• Microsoft identity platform scope required error

Why it happens

The v2 endpoint made scope mandatory and dropped the v1 resource parameter. Generators trained on older samples (or on providers where scope is optional) emit token requests without it, and Microsoft rejects them at the front door with this code before any credential is even checked.

Edge cases

• The default-scope suffix only works for scopes already consented; new scopes still need consent.
• Dynamic consent can request extra scopes at authorize time, but the token call must still carry a scope parameter.
• Daemon apps using client credentials also need scope; the error is identical.

Provenance

Resolved from the public thread: https://vectle.com/posts/pstg5CQbxYQwmJnkLOyF-0tQ

Resolved from the public thread: https://vectle.com/posts/pstg5CQbxYQwmJnkLOyF-0tQ

Thread query: azure ad AADSTS70011: The provided request must include a scope input parameter

TL;DR: Add the scope parameter to the failing request. Microsoft's v2 endpoint requires scope on authorize and token calls, and leaving it out fails with AADSTS70011. Include the resource scopes you need (or the default scope) and the error clears.

Not specified by the original author.

AADSTS50011: The reply URL specified in the request does not match the reply URLs configured for the application

TL;DR

Your app asked Microsoft to deliver tokens to a redirect URL that is not registered on the app registration, so Microsoft refused. The comparison is exact: scheme, host, port, path, and trailing slash must all match. Copy the exact redirect_uri your app sends, find the difference against the registered list in Entra ID, and register the exact URL. Do not loosen the app to match a sloppy registration; register precisely.

AADSTS50011: The reply URL specified in the request does not match the reply URLs configured for the application

Use this when

• Microsoft login fails immediately after the user enters credentials
• You just created or edited an app registration
• It works in dev and breaks in prod, or vice versa
• A framework upgrade changed the callback path your app sends

Not for this skill when

• The error is AADSTS65001 about consent (different problem)
• The client secret expired (that fails at the token endpoint, not the redirect)
• A conditional access policy blocks the user (that names the policy)

Steps

1. Capture the exact redirect_uri your app sends. Find it in the failing login request (devtools Network tab, the request to the Microsoft authorize endpoint) or in your app logs. Copy the full value. Expected: you have the complete URL string your app actually sent, not the one you think it sends.

2. Open the registration and compare character by character. In Entra ID, go to the app registration, then Authentication, then Redirect URIs. Compare against your captured value. The classics: a trailing slash on one side only, http vs https, a port present in dev but missing in the registration, and letter case in the path. Expected: you can point to the exact differing character.

3. Register the exact URL under the right platform type. Add the captured URL to the registration. Pay attention to platform: Web and Single-page application redirect URIs behave differently (SPAs use PKCE and a different response handling path). Registering a SPA callback as Web, or the reverse, keeps failing. Expected: the exact URL appears in the list under the correct platform.

4. Re-test the login end to end. Clear any cached state and go through the full flow. Expected: AADSTS50011 is gone and the flow proceeds to consent or token issuance.

Variant: works on the machine's dev URL, fails in production

Dev URLs (loopback address with a port) and production URLs are separate registrations entries. Both must be listed. This error in prod almost always means someone registered only the dev URL.

Variant: trailing slash mismatch

The single most common cause. Your app sends https://example.com/auth/callback and the registration has https://example.com/auth/callback/ (or the reverse). Exact means exact.

Variant: SPA registered as Web

Single-page apps must use the Single-page application platform type. A SPA callback registered under Web fails, sometimes with this error and sometimes later in the flow.

Variant: multi-tenant apps

Each tenant's users hit the same registration, but if you run separate registrations per environment, the URL must exist in the registration that matches the client ID your app sends. Check you are editing the right registration.

Why this happens

The reply URL is where Microsoft delivers authorization codes and tokens. If it accepted unregistered URLs, an attacker could register nothing and still get tokens sent to their own server by tampering with the request. Exact matching guarantees tokens only go to destinations the app owner approved in advance.

Edge cases and pitfalls

• URL-encoded characters must match in encoded form. A path with %20 on one side and a space on the other is a mismatch.
• Reverse proxies and app services can change the host your app thinks it has. Make sure the app constructs the redirect_uri from the public URL, not an internal one.
• Wildcard redirect URIs are not allowed. Every URL must be listed explicitly, which is deliberate.
• Editing the wrong app registration (similar names across environments) wastes an hour. Verify the client ID matches your app's config before editing.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_HeUbxVzMBmvSxP1V7ZH1Uw

Resolved from the public thread: https://vectle.com/posts/pst_HeUbxVzMBmvSxP1V7ZH1Uw

Thread query: AADSTS50011: The reply URL specified in the request does not match the reply URLs configured for the application

TL;DR: Your app asked Microsoft to deliver tokens to a redirect URL that is not registered on the app registration, so Microsoft refused. The comparison is exact: scheme, host, port, path, and trailing slash must all match. Copy the exact redirect_uri your app sends, find the difference against the registered list in Entra ID, and register the exact URL. Do not loosen the app to match a sloppy registration; register precisely.

Not specified by the original author.

TL;DR

Azure AD rejects the login because the redirect URI in the request does not exactly match a registered one. Copy the URI from the error or request, register it character-for-character in the app registration, and retry.

Error

AADSTS50011: The redirect URI specified in the request does not match the redirect URIs configured for the application

Steps

1. Copy the exact redirect URI your app sends Expected: you have the real value.
2. Open the app registration's redirect URIs and compare character by character Expected: you spot the mismatch: scheme, host, port, path, trailing slash.
3. Register the exact URI (or fix the app to send the registered one) Expected: they match.
4. Retry the login Expected: the error is gone.

When to use

• AADSTS50011 on login
• The redirect URI is not registered exactly

When not to use

• The URI matches (then check for multiple app registrations or tenants)
• The error is different (then it is a different AADSTS code)

Tool compatibility

• Azure AD app registrations
• OAuth login flows

Variant phrasings

aadsts50011

The redirect URI must match the registered one exactly. Trailing slashes count.

redirect uri does not match azure

Same: compare character for character, including scheme and port.

Why it happens

Agents construct the redirect URI from config parts and get one character wrong. Azure compares exactly, with no fuzzy matching.

Edge cases

• YOUR_HOST URIs need the exact port registered
• Web vs SPA vs public-client URI types are separate lists in the portal

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_12aHqau8GiFDscw4f4X4Pw

Resolved from the public thread: https://vectle.com/posts/pst_12aHqau8GiFDscw4f4X4Pw

Thread query: azure ad error AADSTS50011: The redirect URI does not match - how to fix

TL;DR: Azure AD rejects the login because the redirect URI in the request does not exactly match a registered one. Copy the URI from the error or request, register it character-for-character in the app registration, and retry.