fix(tools): close resp.Body on retry cancel and cache http.Client instances (#940)

* fix(tools): close resp.Body on retry cancel and cache http.Client instances

Fix resp.Body leak in DoRequestWithRetry where req.Body (request) was
incorrectly closed instead of resp.Body (response) on context cancel.
Cache http.Client on web search/fetch provider structs and channel
adapters (WeCom, LINE) to avoid per-call allocation overhead.

* fix(channels): preserve original http client timeouts for LINE and WeCom

Split LINE single 60s client into infoClient (10s) for bot info lookups
and apiClient (30s) for messaging API calls. Lower WeCom cached client
base timeout from 60s to 30s (matching uploadMedia), and ensure it is
always >= the configured ReplyTimeout so the per-request context
deadline remains the effective limit.

* refactor(tools): extract timeout consts and deduplicate WebFetchTool constructors

Address PR review feedback from xiaket:
- Define searchTimeout, perplexityTimeout, fetchTimeout, defaultMaxChars,
  and maxRedirects as package-level consts instead of magic numbers.
- Remove misleading "No proxy" comment in NewWebFetchTool.
- Deduplicate NewWebFetchTool by delegating to NewWebFetchToolWithProxy.

* test(utils): add context cancellation test for DoRequestWithRetry

Verify that resp.Body is properly closed when the context is canceled
during retry sleep, covering the C8 resp.Body leak fix.

* fix(utils): close resp in test to satisfy bodyclose linter

* fix(utils): eliminate flakiness in context cancellation retry test

Synchronize cancellation using an onRoundTrip callback from the
transport wrapper instead of a timing-based context timeout. This
ensures the first client.Do completes before cancel fires, so
cancellation always hits during sleepWithCtx.

Author: ItsT0ng

pkg/agent/loop.go

@@ -99,7 +99,7 @@ func registerSharedTools(
 		}
 
 		// Web tools
-		if searchTool := tools.NewWebSearchTool(tools.WebSearchToolOptions{
+		searchTool, err := tools.NewWebSearchTool(tools.WebSearchToolOptions{
 			BraveAPIKey:          cfg.Tools.Web.Brave.APIKey,
 			BraveMaxResults:      cfg.Tools.Web.Brave.MaxResults,
 			BraveEnabled:         cfg.Tools.Web.Brave.Enabled,
@@ -113,10 +113,18 @@ func registerSharedTools(
 			PerplexityMaxResults: cfg.Tools.Web.Perplexity.MaxResults,
 			PerplexityEnabled:    cfg.Tools.Web.Perplexity.Enabled,
 			Proxy:                cfg.Tools.Web.Proxy,
-		}); searchTool != nil {
+		})
+		if err != nil {
+			logger.ErrorCF("agent", "Failed to create web search tool", map[string]any{"error": err.Error()})
+		} else if searchTool != nil {
 			agent.Tools.Register(searchTool)
 		}
-		agent.Tools.Register(tools.NewWebFetchToolWithProxy(50000, cfg.Tools.Web.Proxy))
+		fetchTool, err := tools.NewWebFetchToolWithProxy(50000, cfg.Tools.Web.Proxy)
+		if err != nil {
+			logger.ErrorCF("agent", "Failed to create web fetch tool", map[string]any{"error": err.Error()})
+		} else {
+			agent.Tools.Register(fetchTool)
+		}
 
 		// Hardware tools (I2C, SPI) - Linux only, returns error on other platforms
 		agent.Tools.Register(tools.NewI2CTool())

pkg/channels/line/line.go

@@ -45,11 +45,13 @@ type replyTokenEntry struct {
 type LINEChannel struct {
 	*channels.BaseChannel
 	config         config.LINEConfig
-	botUserID      string   // Bot's user ID
-	botBasicID     string   // Bot's basic ID (e.g. @216ru...)
-	botDisplayName string   // Bot's display name for text-based mention detection
-	replyTokens    sync.Map // chatID -> replyTokenEntry
-	quoteTokens    sync.Map // chatID -> quoteToken (string)
+	infoClient     *http.Client // for bot info lookups (short timeout)
+	apiClient      *http.Client // for messaging API calls
+	botUserID      string       // Bot's user ID
+	botBasicID     string       // Bot's basic ID (e.g. @216ru...)
+	botDisplayName string       // Bot's display name for text-based mention detection
+	replyTokens    sync.Map     // chatID -> replyTokenEntry
+	quoteTokens    sync.Map     // chatID -> quoteToken (string)
 	ctx            context.Context
 	cancel         context.CancelFunc
 }
@@ -69,6 +71,8 @@ func NewLINEChannel(cfg config.LINEConfig, messageBus *bus.MessageBus) (*LINECha
 	return &LINEChannel{
 		BaseChannel: base,
 		config:      cfg,
+		infoClient:  &http.Client{Timeout: 10 * time.Second},
+		apiClient:   &http.Client{Timeout: 30 * time.Second},
 	}, nil
 }
 
@@ -104,8 +108,7 @@ func (c *LINEChannel) fetchBotInfo() error {
 	}
 	req.Header.Set("Authorization", "Bearer "+c.config.ChannelAccessToken)
 
-	client := &http.Client{Timeout: 10 * time.Second}
-	resp, err := client.Do(req)
+	resp, err := c.infoClient.Do(req)
 	if err != nil {
 		return err
 	}
@@ -644,8 +647,7 @@ func (c *LINEChannel) callAPI(ctx context.Context, endpoint string, payload any)
 	req.Header.Set("Content-Type", "application/json")
 	req.Header.Set("Authorization", "Bearer "+c.config.ChannelAccessToken)
 
-	client := &http.Client{Timeout: 30 * time.Second}
-	resp, err := client.Do(req)
+	resp, err := c.apiClient.Do(req)
 	if err != nil {
 		return channels.ClassifyNetError(err)
 	}

pkg/channels/wecom/app.go

@@ -32,6 +32,7 @@ const (
 type WeComAppChannel struct {
 	*channels.BaseChannel
 	config        config.WeComAppConfig
+	client        *http.Client
 	accessToken   string
 	tokenExpiry   time.Time
 	tokenMu       sync.RWMutex
@@ -129,10 +130,18 @@ func NewWeComAppChannel(cfg config.WeComAppConfig, messageBus *bus.MessageBus) (
 		channels.WithReasoningChannelID(cfg.ReasoningChannelID),
 	)
 
+	// Client timeout must be >= the configured ReplyTimeout so the
+	// per-request context deadline is always the effective limit.
+	clientTimeout := 30 * time.Second
+	if d := time.Duration(cfg.ReplyTimeout) * time.Second; d > clientTimeout {
+		clientTimeout = d
+	}
+
 	ctx, cancel := context.WithCancel(context.Background())
 	return &WeComAppChannel{
 		BaseChannel:   base,
 		config:        cfg,
+		client:        &http.Client{Timeout: clientTimeout},
 		ctx:           ctx,
 		cancel:        cancel,
 		processedMsgs: make(map[string]bool),
@@ -306,8 +315,7 @@ func (c *WeComAppChannel) uploadMedia(ctx context.Context, accessToken, mediaTyp
 	}
 	req.Header.Set("Content-Type", writer.FormDataContentType())
 
-	client := &http.Client{Timeout: 30 * time.Second}
-	resp, err := client.Do(req)
+	resp, err := c.client.Do(req)
 	if err != nil {
 		return "", channels.ClassifyNetError(err)
 	}
@@ -364,8 +372,7 @@ func (c *WeComAppChannel) sendImageMessage(ctx context.Context, accessToken, use
 	}
 	req.Header.Set("Content-Type", "application/json")
 
-	client := &http.Client{Timeout: time.Duration(timeout) * time.Second}
-	resp, err := client.Do(req)
+	resp, err := c.client.Do(req)
 	if err != nil {
 		return channels.ClassifyNetError(err)
 	}
@@ -746,8 +753,7 @@ func (c *WeComAppChannel) sendTextMessage(ctx context.Context, accessToken, user
 	}
 	req.Header.Set("Content-Type", "application/json")
 
-	client := &http.Client{Timeout: time.Duration(timeout) * time.Second}
-	resp, err := client.Do(req)
+	resp, err := c.client.Do(req)
 	if err != nil {
 		return channels.ClassifyNetError(err)
 	}

pkg/channels/wecom/bot.go

@@ -25,6 +25,7 @@ import (
 type WeComBotChannel struct {
 	*channels.BaseChannel
 	config        config.WeComConfig
+	client        *http.Client
 	ctx           context.Context
 	cancel        context.CancelFunc
 	processedMsgs map[string]bool // Message deduplication: msg_id -> processed
@@ -93,10 +94,18 @@ func NewWeComBotChannel(cfg config.WeComConfig, messageBus *bus.MessageBus) (*We
 		channels.WithReasoningChannelID(cfg.ReasoningChannelID),
 	)
 
+	// Client timeout must be >= the configured ReplyTimeout so the
+	// per-request context deadline is always the effective limit.
+	clientTimeout := 30 * time.Second
+	if d := time.Duration(cfg.ReplyTimeout) * time.Second; d > clientTimeout {
+		clientTimeout = d
+	}
+
 	ctx, cancel := context.WithCancel(context.Background())
 	return &WeComBotChannel{
 		BaseChannel:   base,
 		config:        cfg,
+		client:        &http.Client{Timeout: clientTimeout},
 		ctx:           ctx,
 		cancel:        cancel,
 		processedMsgs: make(map[string]bool),
@@ -450,8 +459,7 @@ func (c *WeComBotChannel) sendWebhookReply(ctx context.Context, userID, content
 	}
 	req.Header.Set("Content-Type", "application/json")
 
-	client := &http.Client{Timeout: time.Duration(timeout) * time.Second}
-	resp, err := client.Do(req)
+	resp, err := c.client.Do(req)
 	if err != nil {
 		return channels.ClassifyNetError(err)
 	}