diff --git a/apps/server/internal/platform/peertube.go b/apps/server/internal/platform/peertube.go
index 573fab5ec..5527052fb 100644
--- a/apps/server/internal/platform/peertube.go
+++ b/apps/server/internal/platform/peertube.go
@@ -198,6 +198,39 @@ type peertubeChannel struct {
 	DisplayName string `json:"displayName"`
 }
 
+// peerTubeListPageSize is the largest page PeerTube's paginated lists
+// accept. Without a count they stop at 15.
+const peerTubeListPageSize = 100
+
+// listPeerTubePages reads every page of a paginated PeerTube list, which
+// answers {total, data} and is addressed by start and count.
+func listPeerTubePages[T any](ctx context.Context, endpoint, accessToken, label string) ([]T, error) {
+	var items []T
+	for start := 0; ; {
+		var page struct {
+			Total int64 `json:"total"`
+			Data  []T   `json:"data"`
+		}
+		query := url.Values{"count": {strconv.Itoa(peerTubeListPageSize)}, "start": {strconv.Itoa(start)}}
+		body, err := DoRequest(ctx, http.MethodGet, endpoint+"?"+query.Encode(), nil, map[string]string{
+			headerAuthorization: bearerPrefix + accessToken,
+		})
+		if err != nil {
+			return nil, fmt.Errorf("listing peertube %s: %w", label, err)
+		}
+		if err := json.Unmarshal(body, &page); err != nil {
+			return nil, fmt.Errorf("decoding peertube %s: %w", label, err)
+		}
+		items = append(items, page.Data...)
+		start += len(page.Data)
+		// An empty page also ends the list, so a total that overcounts
+		// cannot keep it requesting.
+		if len(page.Data) == 0 || int64(start) >= page.Total {
+			return items, nil
+		}
+	}
+}
+
 func (p *PeerTubeAdapter) listOwnChannels(ctx context.Context, accessToken string) ([]peertubeChannel, error) {
 	me, err := p.fetchMe(ctx, accessToken)
 	if err != nil {
@@ -207,19 +240,7 @@ func (p *PeerTubeAdapter) listOwnChannels(ctx context.Context, accessToken strin
 	if accountName == "" {
 		return nil, fmt.Errorf("peertube account identity is unavailable")
 	}
-	var result struct {
-		Data []peertubeChannel `json:"data"`
-	}
-	body, err := DoRequest(ctx, http.MethodGet, p.instanceURL+"/api/v1/accounts/"+url.PathEscape(accountName)+"/video-channels", nil, map[string]string{
-		headerAuthorization: bearerPrefix + accessToken,
-	})
-	if err != nil {
-		return nil, fmt.Errorf("listing peertube channels: %w", err)
-	}
-	if err := json.Unmarshal(body, &result); err != nil {
-		return nil, fmt.Errorf("decoding peertube channels: %w", err)
-	}
-	return result.Data, nil
+	return listPeerTubePages[peertubeChannel](ctx, p.instanceURL+"/api/v1/accounts/"+url.PathEscape(accountName)+"/video-channels", accessToken, "channels")
 }
 
 func (p *PeerTubeAdapter) ListAccountSelections(ctx context.Context, token *TokenResult) ([]AccountSelectionOption, error) {
@@ -611,18 +632,9 @@ func (p *PeerTubeAdapter) ListComments(ctx context.Context, accessToken, _ strin
 	}
 	// The thread list carries each thread's first comment as a plain comment.
 	// Replies are only returned by the per-thread tree endpoint.
-	var result struct {
-		Data []peertubeComment `json:"data"`
-	}
-	query := url.Values{"count": {strconv.Itoa(peerTubeCommentPageSize)}}
-	body, err := DoRequest(ctx, http.MethodGet, p.instanceURL+"/api/v1/videos/"+url.PathEscape(videoID)+"/comment-threads?"+query.Encode(), nil, map[string]string{
-		headerAuthorization: bearerPrefix + accessToken,
-	})
+	threads, err := listPeerTubePages[peertubeComment](ctx, p.instanceURL+"/api/v1/videos/"+url.PathEscape(videoID)+"/comment-threads", accessToken, "comments")
 	if err != nil {
-		return nil, fmt.Errorf("fetching peertube comments: %w", err)
-	}
-	if err := json.Unmarshal(body, &result); err != nil {
-		return nil, fmt.Errorf("decoding peertube comments: %w", err)
+		return nil, err
 	}
 	comments := make([]Comment, 0)
 	var walk func(node *peertubeCommentNode, parentRef string)
@@ -647,8 +659,8 @@ func (p *PeerTubeAdapter) ListComments(ctx context.Context, accessToken, _ strin
 			walk(&node.Children[i], ref)
 		}
 	}
-	for i := range result.Data {
-		node := peertubeCommentNode{Comment: result.Data[i]}
+	for i := range threads {
+		node := peertubeCommentNode{Comment: threads[i]}
 		if node.Comment.TotalReplies > 0 {
 			tree, err := p.fetchCommentThread(ctx, accessToken, videoID, node.Comment.ID)
 			if err != nil {
diff --git a/apps/server/internal/platform/peertube_test.go b/apps/server/internal/platform/peertube_test.go
index 3d043e232..83c58fffe 100644
--- a/apps/server/internal/platform/peertube_test.go
+++ b/apps/server/internal/platform/peertube_test.go
@@ -153,6 +153,57 @@ func TestPeerTubeLoginAndChannelSelection(t *testing.T) {
 	require.Equal(t, server.URL, selected.InstanceURL)
 }
 
+func TestPeerTubeChannelSelectionReadsEveryChannelPage(t *testing.T) {
+	// GET /api/v1/accounts/{name}/video-channels is paginated: without a
+	// count PeerTube returns 15 channels, and it rejects a count above 100.
+	const channelCount = 105
+	var starts []string
+	mux := http.NewServeMux()
+	mux.HandleFunc("/api/v1/users/me", func(w http.ResponseWriter, _ *http.Request) {
+		_, _ = w.Write([]byte(`{"id":42,"username":"rodrigo","account":{"name":"rodrigo","displayName":"Rodrigo"}}`))
+	})
+	mux.HandleFunc("/api/v1/accounts/rodrigo/video-channels", func(w http.ResponseWriter, r *http.Request) {
+		start, count := 0, 15
+		if value := r.URL.Query().Get("start"); value != "" {
+			start, _ = strconv.Atoi(value)
+		}
+		if value := r.URL.Query().Get("count"); value != "" {
+			count, _ = strconv.Atoi(value)
+		}
+		if count > 100 {
+			http.Error(w, "Should have a number count (max: 100)", http.StatusBadRequest)
+			return
+		}
+		starts = append(starts, strconv.Itoa(start))
+		channels := make([]peertubeChannel, 0, count)
+		for id := start + 1; id <= min(start+count, channelCount); id++ {
+			channels = append(channels, peertubeChannel{ID: int64(id), Name: "channel_" + strconv.Itoa(id), DisplayName: "Channel " + strconv.Itoa(id)})
+		}
+		_ = json.NewEncoder(w).Encode(struct {
+			Total int64             `json:"total"`
+			Data  []peertubeChannel `json:"data"`
+		}{Total: channelCount, Data: channels})
+	})
+	server := httptest.NewServer(mux)
+	defer server.Close()
+
+	adapter := NewPeerTubeAdapter(server.URL)
+	token := &TokenResult{AccessToken: "atok"}
+	options, err := adapter.ListAccountSelections(t.Context(), token)
+	require.NoError(t, err)
+	require.Len(t, options, channelCount)
+	require.Equal(t, []string{"0", "100"}, starts, "channels must be read page by page")
+	require.Equal(t, "channel_105", options[channelCount-1].ID)
+
+	selected, err := adapter.SelectAccount(t.Context(), token, "channel_105")
+	require.NoError(t, err)
+	require.Equal(t, "channel_105", selected.AccountID)
+
+	picker, err := adapter.SearchPublishingOptions(t.Context(), "atok", PublishingOptionsInput{Source: "peertube_channels", Search: "Channel 105"})
+	require.NoError(t, err)
+	require.Equal(t, []DestinationOption{{Value: "channel_105", Label: "Channel 105"}}, picker.Options)
+}
+
 func TestPeerTubeResumableUploadAndPublish(t *testing.T) {
 	server, fake := newFakePeerTube(t)
 	defer server.Close()
@@ -336,6 +387,59 @@ func TestPeerTubeCommentsFetchesTruncatedReplies(t *testing.T) {
 	require.Equal(t, "peertube:video-uuid-1:16", byID["peertube:video-uuid-1:17"].ParentID)
 }
 
+func TestPeerTubeCommentsReadsEveryThreadPage(t *testing.T) {
+	// GET /api/v1/videos/{id}/comment-threads is paginated with start and
+	// count (max 100); total counts threads, newest first.
+	const threadCount = 130
+	var starts []string
+	mux := http.NewServeMux()
+	mux.HandleFunc("/api/v1/videos/video-uuid-1/comment-threads", func(w http.ResponseWriter, r *http.Request) {
+		start, count := 0, 15
+		if value := r.URL.Query().Get("start"); value != "" {
+			start, _ = strconv.Atoi(value)
+		}
+		if value := r.URL.Query().Get("count"); value != "" {
+			count, _ = strconv.Atoi(value)
+		}
+		if count > 100 {
+			http.Error(w, "Should have a number count (max: 100)", http.StatusBadRequest)
+			return
+		}
+		starts = append(starts, strconv.Itoa(start))
+		threads := make([]peertubeComment, 0, count)
+		for index := start; index < min(start+count, threadCount); index++ {
+			id := int64(threadCount - index)
+			threads = append(threads, peertubeComment{ID: id, ThreadID: id, Text: "Thread " + strconv.FormatInt(id, 10), Account: peertubeAccount{Name: "viewer"}})
+		}
+		_ = json.NewEncoder(w).Encode(struct {
+			Total int64             `json:"total"`
+			Data  []peertubeComment `json:"data"`
+		}{Total: threadCount, Data: threads})
+	})
+	server := httptest.NewServer(mux)
+	defer server.Close()
+
+	adapter := NewPeerTubeAdapter(server.URL)
+	comments, err := adapter.ListComments(t.Context(), "token", "channel", "video-uuid-1")
+	require.NoError(t, err)
+	require.Len(t, comments, threadCount)
+	require.Equal(t, []string{"0", "100"}, starts, "threads must be read page by page")
+	require.Equal(t, "peertube:video-uuid-1:1", comments[threadCount-1].ID, "the oldest thread is listed")
+
+	for body, want := range map[string]string{"": "listing peertube comments", "not json": "decoding peertube comments"} {
+		failing := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) {
+			if body == "" {
+				http.Error(w, "unavailable", http.StatusServiceUnavailable)
+				return
+			}
+			_, _ = w.Write([]byte(body))
+		}))
+		_, err := NewPeerTubeAdapter(failing.URL).ListComments(t.Context(), "token", "channel", "video-uuid-1")
+		failing.Close()
+		require.ErrorContains(t, err, want)
+	}
+}
+
 func TestPeerTubeCategoryAndLicencePickers(t *testing.T) {
 	// GET /api/v1/videos/categories and /licences answer with a plain
 	// id-to-label object, not a paginated {"data":[...]} envelope.
diff --git a/changes/peertube-channel-pages.md b/changes/peertube-channel-pages.md
new file mode 100644
index 000000000..9213fb2ee
--- /dev/null
+++ b/changes/peertube-channel-pages.md
@@ -0,0 +1,4 @@
+### Fixed
+
+- PeerTube accounts with more than 15 channels can connect and publish to every channel; the channel list now reads every page instead of stopping at PeerTube's default page of 15.
+- PeerTube videos with more than 100 comment threads bring every thread into the inbox; older threads, and new replies on them, are no longer dropped after the first page.