[Frontend] Add cache_salt support to Anthropic Messages API (#49498)

Signed-off-by: aeon-x <[email protected]>

tests/entrypoints/anthropic/test_anthropic_messages_conversion.py

@@ -1385,3 +1385,26 @@ class TestMessagesFullConverter:
         assert len(result.content) == 1
         assert result.content[0].type == "text"
         assert result.content[0].text == ""
+
+
+# ======================================================================
+# cache_salt pass-through (Issue #46688)
+# ======================================================================
+
+
+class TestCacheSalt:
+    def test_cache_salt_passed_through(self):
+        """cache_salt on the Anthropic request reaches the converted
+        ChatCompletionRequest so prefix-cache isolation works via /v1/messages."""
+        request = _make_request(
+            [{"role": "user", "content": "Hello"}],
+            cache_salt="tenant-abc-secret-salt",
+        )
+        result = _convert(request)
+        assert result.cache_salt == "tenant-abc-secret-salt"
+
+    def test_cache_salt_defaults_to_none(self):
+        """Omitting cache_salt leaves it unset (unchanged default behavior)."""
+        request = _make_request([{"role": "user", "content": "Hello"}])
+        result = _convert(request)
+        assert result.cache_salt is None

vllm/entrypoints/anthropic/protocol.py

@@ -133,6 +133,17 @@ class AnthropicMessagesRequest(BaseModel):
     top_p: float | None = None
 
     # vLLM-specific fields that are not in Anthropic spec
+    cache_salt: str | None = Field(
+        default=None,
+        description=(
+            "If specified, the prefix cache will be salted with the provided "
+            "string to prevent an attacker to guess prompts in multi-user "
+            "environments. The salt should be random, protected from "
+            "access by 3rd parties, and long enough to be "
+            "unpredictable (e.g., 43 characters base64-encoded, corresponding "
+            "to 256 bit)."
+        ),
+    )
     kv_transfer_params: dict[str, Any] | None = Field(
         default=None,
         description="KVTransfer parameters used for disaggregated serving.",

vllm/entrypoints/anthropic/serving.py

@@ -486,6 +486,7 @@ class AnthropicServingMessages(OpenAIServingChat):
             temperature=anthropic_request.temperature,
             top_p=anthropic_request.top_p,
             top_k=anthropic_request.top_k,
+            cache_salt=anthropic_request.cache_salt,
             kv_transfer_params=anthropic_request.kv_transfer_params,
             ec_transfer_params=anthropic_request.ec_transfer_params,
             chat_template_kwargs=anthropic_request.chat_template_kwargs,