sift/nix/module.nix

{ config, lib, pkgs, ... }:
let
  cfg = config.services.sift;
  settingsFormat = pkgs.formats.toml { };
in
{
  options.services.sift = {
    enable = lib.mkEnableOption "sift LLM-oriented fetch and search";

    package = lib.mkOption {
      type = lib.types.package;
      description = "sift package. The flake module sets this to packages.sift.";
    };

    listen = lib.mkOption {
      type = lib.types.str;
      default = "127.0.0.1:8890";
      description = "Address the HTTP server binds to (loopback by default).";
    };

    tailscaleServe = lib.mkOption {
      type = lib.types.bool;
      default = true;
      description = "Advertise the listener on tailscale serve at https://<host>:8443.";
    };

    tailscaleHttpsPort = lib.mkOption {
      type = lib.types.port;
      default = 8443;
      description = "HTTPS port for tailscale serve (443 is SearXNG on orion).";
    };

    denyPrefixes = lib.mkOption {
      type = lib.types.listOf lib.types.str;
      default = [ ];
      description = ''
        Extra CIDRs the SSRF guard blocks at pre-check and connect time.
        Put home WAN /32s here (hairpin into LAN). Set on the host; do not
        commit residential IPs to the sift repo.
      '';
    };

    selfIP = {
      sources = lib.mkOption {
        type = lib.types.listOf lib.types.str;
        default = [ "trace:https://1.1.1.1/cdn-cgi/trace" ];
        description = ''
          Discovery sources in order: ddns:<hostname> and/or
          trace:<url>. Results that are not global unicast are ignored.
        '';
      };
      interval = lib.mkOption {
        type = lib.types.str;
        default = "10m";
        description = "How often to re-discover the WAN address.";
      };
      keepPrevious = lib.mkOption {
        type = lib.types.str;
        default = "24h";
        description = "How long to keep a previous WAN address in the deny set.";
      };
    };

    settings = lib.mkOption {
      type = settingsFormat.type;
      default = { };
      description = "Written to /etc/sift/config.toml. Unused keys are reserved for P1+.";
    };
  };

  config = lib.mkIf cfg.enable {
    assertions = [
      {
        assertion = cfg.package != null;
        message = "services.sift.package must be set (the flake module does this).";
      }
    ];

    users.users.sift = {
      isSystemUser = true;
      group = "sift";
      home = "/var/lib/sift";
    };
    users.groups.sift = { };

    environment.etc."sift/config.toml".source = settingsFormat.generate "sift.toml" (
      {
        listen = cfg.listen;
        self_ip = {
          sources = cfg.selfIP.sources;
          interval = cfg.selfIP.interval;
          keep_previous = cfg.selfIP.keepPrevious;
          path = "/var/lib/sift/self_ip.json";
        };
      }
      // cfg.settings
    );

    # Keys and env secrets are created on-box, never in the store.
    systemd.tmpfiles.rules = [
      "d /var/lib/sift 0750 root sift -"
    ];

    system.activationScripts.sift-secrets = {
      deps = [ "users" ];
      text = ''
        install -d -m 0750 -o root -g sift /var/lib/sift
        if [ ! -f /var/lib/sift/keys.toml ]; then
          umask 027
          printf '%s\n' '# key-hash → { name, workspace }. Generated empty; fill on-box.' > /var/lib/sift/keys.toml
          chown root:sift /var/lib/sift/keys.toml
          chmod 0640 /var/lib/sift/keys.toml
        fi
        if [ ! -f /var/lib/sift/secret.env ]; then
          umask 027
          printf '%s\n' '# BRAVE_API_KEY=  (set on-box; do not read SearXNG secret.env)' > /var/lib/sift/secret.env
          chown root:sift /var/lib/sift/secret.env
          chmod 0640 /var/lib/sift/secret.env
        fi
      '';
    };

    systemd.services.sift = {
      description = "sift LLM-oriented fetch and search";
      after = [ "network-online.target" ];
      wants = [ "network-online.target" ];
      wantedBy = [ "multi-user.target" ];
      serviceConfig = {
        Type = "simple";
        User = "sift";
        Group = "sift";
        ExecStart = "${lib.getExe cfg.package} serve -listen ${cfg.listen} -config /etc/sift/config.toml"
        + lib.concatMapStrings (p: " -deny-prefix ${lib.escapeShellArg p}") cfg.denyPrefixes;
        WorkingDirectory = "/var/lib/sift";
        Restart = "on-failure";
        RestartSec = "5s";
        NoNewPrivileges = true;
        ProtectSystem = "strict";
        ProtectHome = true;
        PrivateTmp = true;
        PrivateDevices = true;
        ProtectKernelTunables = true;
        ProtectKernelModules = true;
        ProtectControlGroups = true;
        ProtectHostname = true;
        ProtectClock = true;
        RestrictAddressFamilies = [
          "AF_UNIX"
          "AF_INET"
          "AF_INET6"
        ];
        RestrictNamespaces = true;
        RestrictRealtime = true;
        RestrictSUIDSGID = true;
        LockPersonality = true;
        MemoryDenyWriteExecute = true;
        MemoryMax = "512M";
        SystemCallArchitectures = "native";
        SystemCallFilter = [
          "@system-service"
          "~@privileged"
        ];
        CapabilityBoundingSet = "";
        AmbientCapabilities = "";
        ReadWritePaths = [ "/var/lib/sift" ];
        ConfigurationDirectory = "sift";
        StateDirectory = "sift";
      };
    };

    systemd.services.tailscale-serve-sift = lib.mkIf cfg.tailscaleServe {
      description = "Advertise sift on Tailscale Serve :${toString cfg.tailscaleHttpsPort}";
      after = [
        "tailscaled.service"
        "network-online.target"
        "sift.service"
      ];
      wants = [
        "tailscaled.service"
        "network-online.target"
      ];
      wantedBy = [ "multi-user.target" ];
      serviceConfig = {
        Type = "oneshot";
        RemainAfterExit = true;
        Restart = "on-failure";
        RestartSec = "5s";
        ExecStart = "${lib.getExe pkgs.tailscale} serve --bg --yes --https=${toString cfg.tailscaleHttpsPort} http://${cfg.listen}";
      };
    };
  };
}