feat: remove cleanup_on_shutdown from semantic cache plugin config

.github/workflows/scripts/validate-helm-config-fields.sh

@@ -637,7 +637,6 @@ bifrost:
         cache_by_model: true
         cache_by_provider: false
         exclude_system_prompt: true
-        cleanup_on_shutdown: true
         vector_store_namespace: "bifrost-cache"
     otel:
       enabled: true
@@ -710,7 +709,6 @@ assert_field_value 'plugins: semantic_cache conversation_history_threshold' '.pl
 assert_field_value 'plugins: semantic_cache cache_by_model' '.plugins.[4].config.cache_by_model' 'true'
 assert_field_value 'plugins: semantic_cache cache_by_provider' '.plugins.[4].config.cache_by_provider' 'false'
 assert_field_value 'plugins: semantic_cache exclude_system_prompt' '.plugins.[4].config.exclude_system_prompt' 'true'
-assert_field_value 'plugins: semantic_cache cleanup_on_shutdown' '.plugins.[4].config.cleanup_on_shutdown' 'true'
 assert_field_value 'plugins: semantic_cache vector_store_namespace' '.plugins.[4].config.vector_store_namespace' '"bifrost-cache"'
 
 # OTEL plugin

core/providers/anthropic/types.go

@@ -1200,18 +1200,18 @@ const (
 type AnthropicToolName string
 
 const (
-	AnthropicToolNameComputer        AnthropicToolName = "computer"
-	AnthropicToolNameWebSearch       AnthropicToolName = "web_search"
-	AnthropicToolNameWebFetch        AnthropicToolName = "web_fetch"
-	AnthropicToolNameBash            AnthropicToolName = "bash"
-	AnthropicToolNameTextEditor      AnthropicToolName = "str_replace_based_edit_tool"
+	AnthropicToolNameComputer   AnthropicToolName = "computer"
+	AnthropicToolNameWebSearch  AnthropicToolName = "web_search"
+	AnthropicToolNameWebFetch   AnthropicToolName = "web_fetch"
+	AnthropicToolNameBash       AnthropicToolName = "bash"
+	AnthropicToolNameTextEditor AnthropicToolName = "str_replace_based_edit_tool"
 	// AnthropicToolNameTextEditorLegacy is the name required for text_editor_20250124
 	// and text_editor_20250429. Newer text_editor_20250728+ use AnthropicToolNameTextEditor.
 	AnthropicToolNameTextEditorLegacy AnthropicToolName = "str_replace_editor"
-	AnthropicToolNameCodeExecution   AnthropicToolName = "code_execution"
-	AnthropicToolNameMemory          AnthropicToolName = "memory"
-	AnthropicToolNameToolSearchBM25  AnthropicToolName = "tool_search_tool_bm25"
-	AnthropicToolNameToolSearchRegex AnthropicToolName = "tool_search_tool_regex"
+	AnthropicToolNameCodeExecution    AnthropicToolName = "code_execution"
+	AnthropicToolNameMemory           AnthropicToolName = "memory"
+	AnthropicToolNameToolSearchBM25   AnthropicToolName = "tool_search_tool_bm25"
+	AnthropicToolNameToolSearchRegex  AnthropicToolName = "tool_search_tool_regex"
 )
 
 type AnthropicToolComputerUse struct {

docs/deployment-guides/helm/plugins.mdx

@@ -178,7 +178,6 @@ Two modes:
 | `bifrost.plugins.semanticCache.config.cache_by_model` | Include model name in cache key | `true` |
 | `bifrost.plugins.semanticCache.config.cache_by_provider` | Include provider name in cache key | `true` |
 | `bifrost.plugins.semanticCache.config.exclude_system_prompt` | Exclude system prompt from cache key | `false` |
-| `bifrost.plugins.semanticCache.config.cleanup_on_shutdown` | Delete cache data on pod shutdown | `false` |
 
 **Semantic mode (with OpenAI embeddings + Weaviate):**
 

docs/features/semantic-caching.mdx

@@ -198,7 +198,6 @@ bifrostConfig := schemas.BifrostConfig{
         "embedding_model": "text-embedding-3-small",
         "dimension": 1536,
 
-        "cleanup_on_shutdown": true,
         "ttl": "5m",
         "threshold": 0.8,
 
@@ -281,7 +280,6 @@ bifrost:
       config:
         dimension: 1
         ttl: "5m"
-        cleanup_on_shutdown: true
         cache_by_model: true
         cache_by_provider: true
 ```
@@ -299,7 +297,6 @@ bifrost:
       "config": {
         "dimension": 1,
         "ttl": "5m",
-        "cleanup_on_shutdown": true,
         "cache_by_model": true,
         "cache_by_provider": true
       }
@@ -671,16 +668,11 @@ The semantic cache automatically handles cleanup to prevent storage bloat:
 - Clear all entries for a cache key
 - Restart Bifrost to clear all cache data
 
-<Warning>
-The semantic cache namespace and all its cache entries are deleted when Bifrost client shuts down **only if `cleanup_on_shutdown` is set to `true`**. By default (`cleanup_on_shutdown: false`), cache data persists between restarts. DO NOT use the plugin's namespace for external purposes.
-</Warning>
-
 <Warning>
 **Dimension / Provider / Model Changes**: A vector store namespace can only hold vectors of **one** dimension. If you change `dimension` (or switch to an embedding `provider`/`model` that produces a different vector size), the existing namespace is **not** recreated automatically — `CreateNamespace` is a no-op when the class/collection already exists. Subsequent writes will fail (vector-size mismatch) and reads will silently miss. Before saving the change, either:
 
 - point `vector_store_namespace` at a fresh name, or
 - drop the existing class/index in your vector store, or
-- set `cleanup_on_shutdown: true` and restart so the old namespace is removed first.
 </Warning>
 
 ---

docs/migration-guides/v1.5.0.mdx

@@ -751,6 +751,56 @@ If you parse `cache_debug` and assumed it was either absent or had `cache_hit: t
 
 ---
 
+## Breaking Change 16: Semantic Cache `cleanup_on_shutdown` Removed
+
+The `cleanup_on_shutdown` option on the semantic cache plugin config has been removed. Cache entries and the vector store namespace are no longer deleted when Bifrost shuts down — cache data always persists between restarts.
+
+**Before:**
+```json
+{
+  "plugins": {
+    "semantic_cache": {
+      "config": {
+        "ttl": "5m",
+        "cleanup_on_shutdown": true
+      }
+    }
+  }
+}
+```
+
+**After:**
+```json
+{
+  "plugins": {
+    "semantic_cache": {
+      "config": {
+        "ttl": "5m"
+      }
+    }
+  }
+}
+```
+
+The field is no longer part of the config schema and will be rejected by validation. Remove it from `config.json`, Helm values, and any `PUT /api/config` payloads.
+
+### How to clear cache data
+
+If you previously relied on `cleanup_on_shutdown: true` to drop the cache on restart, use one of the supported invalidation paths instead:
+
+- `DELETE /api/cache/clear/{cacheId}` — invalidate a single entry
+- `DELETE /api/cache/clear-by-key/{cacheKey}` — invalidate all entries for a cache key
+- Drop the vector store class/collection or point `vector_store_namespace` at a fresh name to start clean
+
+### Dimension / provider / model changes
+
+The previous `cleanup_on_shutdown: true` + restart workflow was the documented escape hatch for changing `dimension` (or switching to an embedding `provider`/`model` that produces a different vector size). That option is gone. To rotate the namespace now, either:
+
+- point `vector_store_namespace` at a fresh name, or
+- drop the existing class/index in your vector store before restarting
+
+---
+
 ## Opting Out: `version: 1` Compatibility Mode
 
 If you are not ready to adopt the new deny-by-default semantics, you can add a single field to `config.json` to restore v1.4.x behavior for all allow-list fields loaded from that file:
@@ -853,6 +903,10 @@ Remove `allow_direct_keys` from `config.json` and any `PUT /api/config` payloads
 <Step title="Switch semantic cache invalidation to cache IDs">
 Replace `DELETE /api/cache/clear/{requestId}` with `DELETE /api/cache/clear/{cacheId}`, and replace `plugin.ClearCacheForRequestID(...)` with `plugin.ClearCacheForCacheID(...)`. Read the cache ID from `extra_fields.cache_debug.cache_id` on the response (now populated on misses too).
 </Step>
+
+<Step title="Remove cleanup_on_shutdown from semantic cache config">
+Drop the `cleanup_on_shutdown` field from the semantic cache plugin config in `config.json`, Helm values, and any API payloads — it is no longer part of the schema. Cache data now always persists across restarts; use the cache clear endpoints or rotate `vector_store_namespace` to drop entries.
+</Step>
 </Steps>
 
 ---

examples/k8s/examples/values-semantic-search-redis.yaml

@@ -18,7 +18,6 @@ bifrost:
         cache_by_model: true
         cache_by_provider: true
         exclude_system_prompt: true
-        cleanup_on_shutdown: false
         vector_store_namespace: "bifrost-semantic-cache"
 
 vectorStore:

examples/k8s/examples/values-semantic-search-weaviate.yaml

@@ -18,7 +18,6 @@ bifrost:
         cache_by_model: true
         cache_by_provider: true
         exclude_system_prompt: false
-        cleanup_on_shutdown: false
         vector_store_namespace: "bifrost-semantic-cache"
 
 vectorStore:

helm-charts/bifrost/templates/_helpers.tpl

@@ -1100,9 +1100,6 @@ false
 {{- if hasKey $inputConfig "exclude_system_prompt" }}
 {{- $_ := set $scConfig "exclude_system_prompt" $inputConfig.exclude_system_prompt }}
 {{- end }}
-{{- if hasKey $inputConfig "cleanup_on_shutdown" }}
-{{- $_ := set $scConfig "cleanup_on_shutdown" $inputConfig.cleanup_on_shutdown }}
-{{- end }}
 {{- $plugin := dict "enabled" true "name" "semantic_cache" "config" $scConfig }}
 {{- if hasKey .Values.bifrost.plugins.semanticCache "version" }}{{- $_ := set $plugin "version" (.Values.bifrost.plugins.semanticCache.version | int) }}{{- end }}
 {{- $plugins = append $plugins $plugin }}

helm-charts/bifrost/values.schema.json

@@ -818,9 +818,6 @@
                     "exclude_system_prompt": {
                       "type": "boolean"
                     },
-                    "cleanup_on_shutdown": {
-                      "type": "boolean"
-                    },
                     "vector_store_namespace": {
                       "type": "string"
                     },

helm-charts/bifrost/values.yaml

@@ -420,7 +420,6 @@ bifrost:
         cache_by_model: true
         cache_by_provider: true
         exclude_system_prompt: false
-        cleanup_on_shutdown: false
         vector_store_namespace: ""
 
     otel:

plugins/semanticcache/config_unmarshal_test.go

@@ -43,7 +43,6 @@ func TestUnmarshalJSON_AllFields(t *testing.T) {
 	input := `{
 		"provider": "openai",
 		"embedding_model": "text-embedding-3-small",
-		"cleanup_on_shutdown": true,
 		"dimension": 1536,
 		"ttl": "10m",
 		"threshold": 0.9,
@@ -66,9 +65,6 @@ func TestUnmarshalJSON_AllFields(t *testing.T) {
 	if config.EmbeddingModel != "text-embedding-3-small" {
 		t.Errorf("EmbeddingModel: expected %q, got %q", "text-embedding-3-small", config.EmbeddingModel)
 	}
-	if !config.CleanUpOnShutdown {
-		t.Error("CleanUpOnShutdown: expected true")
-	}
 	if config.Dimension != 1536 {
 		t.Errorf("Dimension: expected 1536, got %d", config.Dimension)
 	}
@@ -136,32 +132,32 @@ func TestUnmarshalJSON_TTLFormats(t *testing.T) {
 
 func TestUnmarshalJSON_BoolPointerFields(t *testing.T) {
 	tests := []struct {
-		name                string
-		json                string
-		expectCacheByModel  *bool
-		expectCacheByProv   *bool
-		expectExcludeSys    *bool
+		name               string
+		json               string
+		expectCacheByModel *bool
+		expectCacheByProv  *bool
+		expectExcludeSys   *bool
 	}{
 		{
-			name:                "all set to true",
-			json:                `{"dimension": 1536, "cache_by_model": true, "cache_by_provider": true, "exclude_system_prompt": true}`,
-			expectCacheByModel:  bifrost.Ptr(true),
-			expectCacheByProv:   bifrost.Ptr(true),
-			expectExcludeSys:    bifrost.Ptr(true),
+			name:               "all set to true",
+			json:               `{"dimension": 1536, "cache_by_model": true, "cache_by_provider": true, "exclude_system_prompt": true}`,
+			expectCacheByModel: bifrost.Ptr(true),
+			expectCacheByProv:  bifrost.Ptr(true),
+			expectExcludeSys:   bifrost.Ptr(true),
 		},
 		{
-			name:                "all set to false",
-			json:                `{"dimension": 1536, "cache_by_model": false, "cache_by_provider": false, "exclude_system_prompt": false}`,
-			expectCacheByModel:  bifrost.Ptr(false),
-			expectCacheByProv:   bifrost.Ptr(false),
-			expectExcludeSys:    bifrost.Ptr(false),
+			name:               "all set to false",
+			json:               `{"dimension": 1536, "cache_by_model": false, "cache_by_provider": false, "exclude_system_prompt": false}`,
+			expectCacheByModel: bifrost.Ptr(false),
+			expectCacheByProv:  bifrost.Ptr(false),
+			expectExcludeSys:   bifrost.Ptr(false),
 		},
 		{
-			name:                "all omitted",
-			json:                `{"dimension": 1536}`,
-			expectCacheByModel:  nil,
-			expectCacheByProv:   nil,
-			expectExcludeSys:    nil,
+			name:               "all omitted",
+			json:               `{"dimension": 1536}`,
+			expectCacheByModel: nil,
+			expectCacheByProv:  nil,
+			expectExcludeSys:   nil,
 		},
 	}
 

plugins/semanticcache/main.go

@@ -31,7 +31,6 @@ type Config struct {
 	EmbeddingModel string                `json:"embedding_model,omitempty"` // Model to use for generating embeddings (optional)
 
 	// Plugin behavior settings
-	CleanUpOnShutdown    bool          `json:"cleanup_on_shutdown,omitempty"`    // Clean up cache on shutdown (default: false)
 	TTL                  time.Duration `json:"ttl,omitempty"`                    // Time-to-live for cached responses (default: 5min)
 	Threshold            float64       `json:"threshold,omitempty"`              // Cosine similarity threshold for semantic matching (0 = unset → default 0.8)
 	VectorStoreNamespace string        `json:"vector_store_namespace,omitempty"` // Namespace for vector store (optional)
@@ -684,43 +683,6 @@ func (plugin *Plugin) Cleanup() error {
 	// and stopCh is still in memory. This call evicts those before we return.
 	plugin.cleanupOldStreamAccumulators()
 
-	// Only clean up cache entries if configured to do so
-	if !plugin.config.CleanUpOnShutdown {
-		plugin.logger.Debug("Cleanup on shutdown is disabled, skipping cache cleanup")
-		return nil
-	}
-
-	// Clean up all cache entries created by this plugin
-	ctx, cancel := context.WithTimeout(context.Background(), CacheSetTimeout)
-	defer cancel()
-
-	plugin.logger.Debug("Starting cleanup of cache entries...")
-
-	// Delete all cache entries created by this plugin
-	queries := []vectorstore.Query{
-		{
-			Field:    "from_bifrost_semantic_cache_plugin",
-			Operator: vectorstore.QueryOperatorEqual,
-			Value:    true,
-		},
-	}
-
-	results, err := plugin.store.DeleteAll(ctx, plugin.config.VectorStoreNamespace, queries)
-	if err != nil {
-		return fmt.Errorf("failed to delete cache entries: %w", err)
-	}
-
-	for _, result := range results {
-		if result.Status == vectorstore.DeleteStatusError {
-			plugin.logger.Warn("Failed to delete cache entry: %s", result.Error)
-		}
-	}
-	plugin.logger.Debug("Cleanup completed - deleted all cache entries")
-
-	if err := plugin.store.DeleteNamespace(ctx, plugin.config.VectorStoreNamespace); err != nil {
-		return fmt.Errorf("failed to delete namespace: %w", err)
-	}
-
 	return nil
 }