Secure Azure Functions with JWT access tokens

I have completely rewritten this post. You can find the original post here.

Out of the box it is only possible to secure your Azure Functions via Function Keys (API-Keys), which sometimes might not fit into your requirements. When using HttpTrigger we luckily have access to the current request and are therefor able to implement our own custom authentication/authorization methods. You could for example use JWT access tokens issued by an OpenID provider to control authentication/authorization.

In this post you learn how to validate JWT access tokens and controlling access to your Azure Function.

First lets see the function.

public static class HelloWorldFunction
    public static async Task<HttpResponseMessage> Run(
        [HttpTrigger(AuthorizationLevel.Anonymous, "get", Route = null)]HttpRequestMessage req)
        // Authentication boilerplate code start
        ClaimsPrincipal principal;
        if ((principal = await Security.ValidateTokenAsync(req.Headers.Authorization)) == null)
            return req.CreateResponse(HttpStatusCode.Unauthorized);
        // Authentication boilerplate code end

        return req.CreateResponse(HttpStatusCode.OK, "Hello " + principal.Identity.Name);

It is very important that you set the authorization level to anonymous, since we want to skip all checks done by Azure Functions. Then we need to add the “authentication boilerplate code” to every function, we want to protect with JWT access tokens. Unfortunately there is currently no generic way to add this, e.g. via attributes.

Now we need to implement the validation method.

using Microsoft.IdentityModel.Protocols;
using Microsoft.IdentityModel.Protocols.OpenIdConnect;
using Microsoft.IdentityModel.Tokens;
using System;
using System.IdentityModel.Tokens.Jwt;
using System.Net.Http.Headers;
using System.Security.Claims;
using System.Threading;
using System.Threading.Tasks;

namespace FunctionApp5
    public static class Security
        private static readonly IConfigurationManager<OpenIdConnectConfiguration> _configurationManager;

        static Security()
            var issuer = Environment.GetEnvironmentVariable("ISSUER");

            var documentRetriever = new HttpDocumentRetriever();
            documentRetriever.RequireHttps = issuer.StartsWith("https://");

            _configurationManager = new ConfigurationManager<OpenIdConnectConfiguration>(
                new OpenIdConnectConfigurationRetriever(),

        public static async Task<ClaimsPrincipal> ValidateTokenAsync(AuthenticationHeaderValue value)
            if (value?.Scheme != "Bearer")
                return null;

            var config = await _configurationManager.GetConfigurationAsync(CancellationToken.None);
            var issuer = Environment.GetEnvironmentVariable("ISSUER");
            var audience = Environment.GetEnvironmentVariable("AUDIENCE");

            var validationParameter = new TokenValidationParameters()
                RequireSignedTokens = true,
                ValidAudience = audience,
                ValidateAudience = true,
                ValidIssuer = issuer,
                ValidateIssuer = true,
                ValidateIssuerSigningKey = true,
                ValidateLifetime = true,
                IssuerSigningKeys = config.SigningKeys

            ClaimsPrincipal result = null;
            var tries = 0;

            while (result == null && tries <= 1)
                    var handler = new JwtSecurityTokenHandler();
                    result = handler.ValidateToken(value.Parameter, validationParameter, out var token);
                catch (SecurityTokenSignatureKeyNotFoundException)
                    // This exception is thrown if the signature key of the JWT could not be found.
                    // This could be the case when the issuer changed its signing keys, so we trigger a 
                    // refresh and retry validation.
                catch (SecurityTokenException)
                    return null;

            return result;

Here we use a ConfigurationManager in order to retrieve the signing keys from the OpenID provider. It is very important to have the ConfigurationManager as a static, since it is caching the configuration in order to reduce HTTP requests to the OpenID provider. The ConfigurationManager refreshes the configuration after configurable timeouts. Next in the ValidateTokenAsync method we setup TokenValidationParameter. Here we want to enable as much validation as possible.

Once we configured the validation parameters, we call ValidateToken. If the token can not be validated a SecurityTokenException is thrown. There are several derived exceptions, but in this case we just care for the SecurityTokenSignatureKeyNotFoundException exception, which is thrown if the JwtSecurityTokenHandler can not find the signature keys, that are specified in the JWT. This could be the case, when the issuer changed its signature keys, after the ConfigurationManager fetched the configuration. Therefore we trigger a refresh on the ConfigurationManger and retry to validate the JWT.

If the validation is successful we return a ClaimsPrincipal which contains the claims provided by the token.

If you are not using OpenID you need to change the ConfigurationManager options. Alternatively you can remove the ConfigurationManger and provide static signing keys via the TokenValidationParameters.