Files
HomeAgent/internal/plugins/remotedevice/plugin.go
JianFeeeee ba5785036a feat: 设备鉴权迁移至客户端 + 插件卸载保护
安全修复(客户端鉴权):
- remotedevice 服务端移除授权状态存储(authorized map/SetAuthorized/handleDeviceAuth)
- DeviceMeta.Authorized 改为设备 hello 自报,服务端仅透传展示
- device_ctl_* 工具移除服务端授权检查,无条件转发,设备端自行决定是否执行
- 共享设备桥库 Bridge 新增本地 authorized 状态,未授权收到 cmd 直接拒绝
- waiter: --device-authorized / device_authorized 配置控制本地授权
- GUI: 授权存 gui-prefs 本地文件;设备页仅本机可切换开关
- webui /device/auth 旧路径返回 410 Gone
- 根因:agent 可经 config_set 篡改服务端授权配置自行授权设备

插件管理强化:
- 内置插件禁止卸载(IsBuiltinPlugin + 409),外部插件卸载即时生效
- 卸载不存在插件返回 404;移除误导性 reload_required 提示
- webui 插件路由:名称白名单校验防路径穿越、保留字路径保护
2026-08-24 19:26:11 +08:00

295 lines
9.5 KiB
Go
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

package remotedevice
import (
"context"
"crypto/rand"
"encoding/hex"
"encoding/json"
"fmt"
"log"
"net/http"
"strings"
"sync"
"time"
"gitcode.com/JianFeeeee/HomeAgent/internal/plugin"
sdk "gitcode.com/JianFeeeee/HomeAgent/internal/sdk"
)
func init() {
plugin.RegisterPluginMeta("remotedevice", "远程设备网关", "Remote Device")
plugin.RegisterFactory("remotedevice", func(name string, config map[string]interface{}) (sdk.Plugin, error) {
return New(name), nil
})
}
const defaultAddr = "127.0.0.1:9890"
// Plugin 是 remotedevice 设备接入网关插件:
// 持有 HTTP 服务WS 设备通道 + REST 管理面)与 devicectl Deviceagent 工具)。
type Plugin struct {
name string
registry *Registry
mux *http.ServeMux
server *http.Server
addr string
token string
sdk *sdk.PluginSDK
dev *devicectlDevice
}
func New(name string) *Plugin {
return &Plugin{
name: name,
registry: NewRegistry(),
mux: http.NewServeMux(),
}
}
func (p *Plugin) Name() string { return p.name }
func genToken() string {
buf := make([]byte, 16)
if _, err := rand.Read(buf); err != nil {
return fmt.Sprintf("tok-%d", time.Now().UnixNano())
}
return hex.EncodeToString(buf)
}
func newReqID() string {
buf := make([]byte, 8)
if _, err := rand.Read(buf); err != nil {
return fmt.Sprintf("req-%d", time.Now().UnixNano())
}
return hex.EncodeToString(buf)
}
func (p *Plugin) Start(s *sdk.PluginSDK) error {
s.SetAutoRestart(true)
p.sdk = s
// ---- 设置 ----------------
s.Settings().RegisterDef(sdk.ConfigDef{Key: "listen_addr", Default: defaultAddr, Type: "string", DisplayName: "监听地址", Description: "设备网关 HTTP/WS 监听地址(默认 127.0.0.1:9890仅本机", Category: "remotedevice"})
s.Settings().RegisterDef(sdk.ConfigDef{Key: "ws_token", Default: "", Type: "password", DisplayName: "接入 Token", Description: "设备绑定/接入时使用的令牌;留空启动时自动生成", Category: "remotedevice"})
// 注意:不注册 authorized_devices 设置项 —— 鉴权在设备端执行(客户端存储),
// 服务端不保存授权状态,避免 agent 经 config_set 工具自行授权。
p.addr = defaultAddr
if v, _ := s.Settings().Get("listen_addr"); v != nil {
if s2, ok := v.(string); ok && s2 != "" {
p.addr = s2
}
}
p.token = ""
if v, _ := s.Settings().Get("ws_token"); v != nil {
if s2, ok := v.(string); ok && s2 != "" {
p.token = s2
}
}
if p.token == "" {
p.token = genToken()
if err := s.Settings().Set("ws_token", p.token); err != nil {
log.Printf("[remotedevice] persist ws_token: %v", err)
}
}
p.registry.SetAcceptToken(func(provided string) bool {
return provided != "" && provided == p.token
})
// ---- devicectl Deviceagent 工具) ----------------
p.dev = &devicectlDevice{reg: p.registry}
// screensee 视觉描述回调:截屏回传后用视觉模型描述屏幕内容
p.dev.SetSeeHandler(p.describeScreen)
if err := s.RegisterChannel("devicectl", p.dev); err != nil {
log.Printf("[remotedevice] register devicectl channel: %v", err)
}
// ---- 设备主动上报事件 → agent 注入 ----------------
// 摄像头发现异常/传感器报警等场景:设备经 WS op=event 上报,
// 插件将其格式化为文本经 SDK InjectText 异步注入 agentsource=device/{id}
// 回复路由回 device/{id} 通道),同时发 EventBus 供 WebUI 展示。
// 节流:同设备同类型事件 10s 内去重,防传感器风暴。
lastEventAt := map[string]time.Time{}
var eventMu sync.Mutex
p.registry.SetEventHandler(func(deviceID string, msg map[string]interface{}) {
evtType, _ := msg["type"].(string)
if evtType == "" {
evtType = "unknown"
}
key := deviceID + "|" + evtType
eventMu.Lock()
if last, ok := lastEventAt[key]; ok && time.Since(last) < 10*time.Second {
eventMu.Unlock()
log.Printf("[remotedevice] event throttled: %s from %s", evtType, deviceID)
return
}
lastEventAt[key] = time.Now()
eventMu.Unlock()
// 组装人类可读的事件文本agent 可直接理解)
detail, _ := msg["detail"].(string)
if detail == "" {
if d, ok := msg["payload"].(map[string]interface{}); ok {
b, _ := json.Marshal(d)
detail = string(b)
}
}
text := fmt.Sprintf("【设备事件上报】设备 %s 触发事件 %s", deviceID, evtType)
if detail != "" {
text += "" + detail
}
text += "。请关注此事件并按需处理(如通知用户、调用相关工具核实)。"
log.Printf("[remotedevice] event from %s: %s", deviceID, evtType)
if p.sdk != nil {
// 异步注入:不阻塞 WS 读循环;回复路由回 device/{id} 输出通道
p.sdk.InjectInput("device/"+deviceID, "device/"+deviceID, "text", map[string]interface{}{"content": text})
}
})
// ---- REST 管理面 + WS 设备通道 ----------------
p.registerRoutes()
p.server = &http.Server{Addr: p.addr, Handler: p.mux}
go func() {
log.Printf("[remotedevice] device gateway listening on %s", p.addr)
if err := p.server.ListenAndServe(); err != nil && err != http.ErrServerClosed {
log.Printf("[remotedevice] server error: %v", err)
}
}()
return nil
}
func (p *Plugin) registerRoutes() {
// 设备通道WS
p.mux.HandleFunc("/api/v1/device/ws", p.registry.ServeWS)
// REST 管理面(全部需 token
// 注意:/api/v1/device/auth 已移除 —— 授权由设备端控制,服务端不提供授权接口。
p.mux.HandleFunc("/api/v1/device", p.requireToken(p.handleDeviceList))
p.mux.HandleFunc("/api/v1/device/online", p.requireToken(p.handleDeviceOnline))
p.mux.HandleFunc("/api/v1/device/", p.requireToken(p.handleDeviceByID))
p.mux.HandleFunc("/api/v1/device/push", p.requireToken(p.handleDevicePush))
}
// requireToken 校验 REST 请求的接入令牌X-API-Key header 或 ?token=)。
func (p *Plugin) requireToken(next http.HandlerFunc) http.HandlerFunc {
return func(w http.ResponseWriter, r *http.Request) {
k := r.Header.Get("X-API-Key")
if k == "" {
k = r.URL.Query().Get("token")
}
if k == "" || k != p.token {
http.Error(w, "unauthorized", http.StatusUnauthorized)
return
}
next(w, r)
}
}
func writeJSON(w http.ResponseWriter, status int, v interface{}) {
w.Header().Set("Content-Type", "application/json")
w.WriteHeader(status)
_ = json.NewEncoder(w).Encode(v)
}
func (p *Plugin) handleDeviceList(w http.ResponseWriter, r *http.Request) {
if r.Method != http.MethodGet {
http.NotFound(w, r)
return
}
writeJSON(w, http.StatusOK, map[string]interface{}{"devices": p.registry.List()})
}
func (p *Plugin) handleDeviceOnline(w http.ResponseWriter, r *http.Request) {
if r.Method != http.MethodGet {
http.NotFound(w, r)
return
}
writeJSON(w, http.StatusOK, map[string]interface{}{"devices": p.registry.OnlineList()})
}
func (p *Plugin) handleDeviceByID(w http.ResponseWriter, r *http.Request) {
id := strings.TrimPrefix(r.URL.Path, "/api/v1/device/")
if id == "" || strings.Contains(id, "/") {
http.NotFound(w, r)
return
}
if r.Method == http.MethodGet {
m, ok := p.registry.Get(id)
if !ok {
writeJSON(w, http.StatusNotFound, map[string]interface{}{"error": "device not found"})
return
}
writeJSON(w, http.StatusOK, m)
return
}
http.NotFound(w, r)
}
func (p *Plugin) handleDevicePush(w http.ResponseWriter, r *http.Request) {
if r.Method != http.MethodPost {
http.NotFound(w, r)
return
}
var req struct {
DeviceID string `json:"device_id"`
Payload map[string]interface{} `json:"payload"`
}
if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
writeJSON(w, http.StatusBadRequest, map[string]interface{}{"error": err.Error()})
return
}
if req.DeviceID == "" {
writeJSON(w, http.StatusBadRequest, map[string]interface{}{"error": "device_id required"})
return
}
if err := p.registry.PushJSON(req.DeviceID, req.Payload); err != nil {
writeJSON(w, http.StatusNotFound, map[string]interface{}{"error": err.Error()})
return
}
writeJSON(w, http.StatusOK, map[string]interface{}{"status": "ok"})
}
// describeScreen 用视觉模型描述设备屏幕截图screensee 回调)。
// provider 为空时使用默认 LLM 源;模型不支持视觉时返回友好错误。
func (p *Plugin) describeScreen(dataURL string, provider string) string {
if p.sdk == nil || p.sdk.LLM() == nil {
return "LLM 不可用,无法描述屏幕内容"
}
llm := p.sdk.LLM()
req := &sdk.LLMCompletionRequest{
MaxTokens: 2048,
Messages: []sdk.LLMMessage{{
Role: "user",
Blocks: []sdk.LLMContentBlock{
{Type: "text", Text: "这是用户设备的屏幕截图。请详细描述屏幕上显示的内容:正在运行的窗口/应用、可见的文字内容、界面状态等。如果是代码编辑器或终端,尽量转述关键文字信息。"},
{Type: "image_url", ImageURL: dataURL},
},
}},
}
// 指定源:临时切换(低频操作,用完恢复原源)
if provider != "" {
prev := llm.CurrentSource()
if err := llm.SetSource(provider); err != nil {
log.Printf("[remotedevice] screensee set source %s: %v", provider, err)
} else if prev != "" {
defer func() { _ = llm.SetSource(prev) }()
}
}
ctx, cancel := context.WithTimeout(context.Background(), 120*time.Second)
defer cancel()
resp, err := llm.Chat(ctx, req)
if err != nil {
return fmt.Sprintf("屏幕截图视觉描述失败: %v当前模型可能不支持图像输入", err)
}
return resp.Content
}
func (p *Plugin) Stop() error {
if p.server != nil {
ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
defer cancel()
return p.server.Shutdown(ctx)
}
return nil
}