feat(gitea): add Anubis anti-crawler proxy

Anubis (https://anubis.techaro.lol) is a lightweight proof-of-work
challenge that protects web services from AI crawlers and scrapers.

home/baseline/ssh.nix

@@ -13,25 +13,23 @@
     includes = [ "config.local" ];
     enableDefaultConfig = false;
 
-    matchBlocks =
+    settings =
       let
         nas = {
-          user = "admin";
+          User = "admin";
         };
         owned = {
-          user = "greg";
+          User = "greg";
         };
       in
       {
         inherit nas;
 
         "*" = {
-          dynamicForwards = [ { port = 10240; } ];
-          serverAliveInterval = 60;
-          extraOptions = {
-            LogLevel = "error";
-            SetEnv = "TERM=xterm-256color";
-          };
+          DynamicForward = [ "10240" ];
+          ServerAliveInterval = 60;
+          LogLevel = "error";
+          SetEnv = { TERM = "xterm-256color"; };
         };
 
         "10.42.1.4" = lib.hm.dag.entryBefore [ "10.42.*" ] nas;
@@ -41,46 +39,46 @@
         "chronicles.thehellings.lan" = lib.hm.dag.entryBefore [ "*.thehellings.lan" ] nas;
 
         gh = {
-          user = "git";
-          hostname = "github.com";
+          User = "git";
+          Hostname = "github.com";
         };
         "src" = {
-          user = "git";
-          hostname = "jeremiah.shire-zebra.ts.net";
-          port = 32222;
+          User = "git";
+          Hostname = "jeremiah.shire-zebra.ts.net";
+          Port = 32222;
         };
         srcpub = {
-          user = "git";
-          hostname = "src.thehellings.com";
-          port = 2222;
+          User = "git";
+          Hostname = "src.thehellings.com";
+          Port = 2222;
         };
         ivr = {
-          user = "git";
-          hostname = "gitlab.com";
+          User = "git";
+          Hostname = "gitlab.com";
         };
 
         "ivr.thehellings.lan" = lib.hm.dag.entryBefore [ "ivr" ] {
-          user = "gregory.hellings";
+          User = "gregory.hellings";
         };
 
         "*.thehellings.lan" = owned;
         "10.42.*" = owned;
 
         "host.crosswire.org crosswire" = {
-          hostname = "host.crosswire.org";
-          user = "ghellings";
+          Hostname = "host.crosswire.org";
+          User = "ghellings";
         };
 
         fedpeople = {
-          hostname = "fedorapeople.org";
-          user = "greghellings";
+          Hostname = "fedorapeople.org";
+          User = "greghellings";
         };
 
         "src.fedoraproject.org pkgs.fedoraproject.org" = {
-          user = "greghellings";
+          User = "greghellings";
         };
 
-        "127.*".extraOptions = {
+        "127.*" = {
           PubkeyAcceptedAlgorithms = "+ssh-rsa";
           HostkeyAlgorithms = "+ssh-rsa";
         };

hosts/unstable/linode/git.nix

@@ -1,4 +1,4 @@
-{ ... }:
+{ config, ... }:
 
 let
   srcDomain = "src.thehellings.com";
@@ -6,15 +6,23 @@ let
 in
 {
   greg.proxies."${srcDomain}" = {
-    target = "https://gitea.shire-zebra.ts.net";
+    target = "http://unix:${config.services.anubis.instances.git.settings.BIND}";
     ssl = true;
     genAliases = false;
     extraConfig = ''
-      proxy_ssl_verify off;
-      proxy_ssl_server_name on;
-      proxy_set_header X-Forwarded-Proto $scheme;
-      proxy_set_header X-Forwarded-Ssl on;
+      #proxy_ssl_verify off;
+      #proxy_ssl_server_name on;
+
+      proxy_set_header X-Real-IP $remote_addr;
+      proxy_set_header X-Http-Version $server_protocol;
+      proxy_set_header User-Agent $http_user_agent;
       client_max_body_size 100000m;
+
+      #proxy_set_header Host $host;
+      #proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+      #proxy_set_header X-Forwarded-Proto $scheme;
+      #proxy_set_header X-Forwarded-Ssl on;
+
       # Ultimate AI Block List v1.7 20250924
       # https://perishablepress.com/ultimate-ai-block-list/
 
@@ -87,6 +95,44 @@ in
 
   networking.firewall.allowedTCPPorts = [ sshPort ];
 
+  services = {
+    anubis = {
+      instances = {
+        git = {
+          enable = true;
+          settings = {
+            BIND = "/run/anubis/anubis-git/anubis.sock";
+            COOKIE_DOMAIN = "thehellings.com";
+            SERVE_ROBOTS_TXT = true;
+            TARGET = "https://gitea.shire-zebra.ts.net/";
+          };
+        };
+      };
+    };
+
+    haproxy = {
+      enable = true;
+      config = ''
+        global
+          daemon
+          maxconn 20
+
+        defaults
+          timeout connect 500s
+          timeout client 500s
+          timeout server 1h
+
+        listen gitsshd
+          bind *:${toString sshPort}
+          timeout client 1h
+          mode tcp
+          server git-isaiah isaiah.shire-zebra.ts.net:32222
+          server git-jeremiah jeremiah.shire-zebra.ts.net:32222
+          server git-zeke zeke.shire-zebra.ts.net:32222
+      '';
+    };
+  };
+
   systemd.services = {
     haproxy = {
       after = [
@@ -98,25 +144,5 @@ in
     };
   };
 
-  services.haproxy = {
-    enable = true;
-    config = ''
-      global
-        daemon
-        maxconn 20
-
-      defaults
-        timeout connect 500s
-        timeout client 500s
-        timeout server 1h
-
-      listen gitsshd
-        bind *:${toString sshPort}
-        timeout client 1h
-        mode tcp
-        server git-isaiah isaiah.shire-zebra.ts.net:32222
-        server git-jeremiah jeremiah.shire-zebra.ts.net:32222
-        server git-zeke zeke.shire-zebra.ts.net:32222
-    '';
-  };
+  users.users.nginx.extraGroups = [ config.users.groups.anubis.name ];
 }

manifests/gitea/chart.yaml

@@ -47,12 +47,6 @@ spec:
         type: LoadBalancer
         port: 32222
         nodePort: 32222
-      # Anubis sidecar port — ingress routes here; Anubis proxies to :3000
-      additionalPorts:
-        - name: anubis
-          port: 8080
-          targetPort: 8080
-          protocol: TCP
 
     gitea:
       admin:
@@ -139,14 +133,16 @@ spec:
             value: "4"
           - name: METRICS_BIND
             value: ":9090"
+          - name: POLICY_FNAME
+            value: "/etc/anubis/policy.yaml"
           - name: SERVE_ROBOTS_TXT
             value: "true"
           - name: TARGET
             value: "http://localhost:3000"
-          - name: OG_PASSTHROUGH
-            value: "true"
-          - name: OG_EXPIRY_TIME
-            value: "24h"
+        volumeMounts:
+          - name: anubis-policy
+            mountPath: /etc/anubis
+            readOnly: true
         resources:
           limits:
             cpu: "750m"
@@ -165,6 +161,11 @@ spec:
           seccompProfile:
             type: RuntimeDefault
 
+    extraVolumes:
+      - name: anubis-policy
+        configMap:
+          name: anubis-policy
+
     strategy:
       type: RollingUpdate
       rollingUpdate:

manifests/gitea/ingress.yaml

@@ -8,8 +8,7 @@ spec:
     service:
       name: gitea-release-http
       port:
-        # Route through Anubis anti-crawler sidecar instead of directly to Gitea
-        name: anubis
+        name: http
   tls:
     - hosts:
         - gitea

manifests/gitea/secrets.yaml

@@ -15,8 +15,8 @@ spec:
         storage: |-
           MINIO_ACCESS_KEY_ID={{ .minio_nas1_key }}
           MINIO_SECRET_ACCESS_KEY={{ .minio_nas1_secret }}
-        minio_key: "{{ .minio_nas1_key }}"
-        minio_secret: "{{ .minio_nas1_secret }}"
+        #minio_key: "{{ .minio_nas1_key }}"
+        #minio_secret: "{{ .minio_nas1_secret }}"
   secretStoreRef:
     name: bitwarden-login
     kind: ClusterSecretStore