Xipong
0543fa2f1b
Third-party Anthropic-compatible relays serving a DeepSeek thinking model (deepseek-r/v4/pro/flash) follow the same replay contract as DeepSeek's native /anthropic endpoint: strip signed thinking blocks, preserve unsigned ones. Previously these sessions hit the generic third-party path and lost ALL thinking blocks, breaking cross-turn reasoning coherence. Detection is model-name based (vendor prefixes stripped), gated on the existing _is_third_party check so direct Anthropic traffic is untouched.
153 lines
6.7 KiB
Python
153 lines
6.7 KiB
Python
"""Regression guard: preserve thinking blocks on DeepSeek's /anthropic endpoint.
|
|
|
|
DeepSeek's ``api.deepseek.com/anthropic`` route speaks the Anthropic Messages
|
|
protocol but, when thinking mode is enabled, requires ``thinking`` blocks from
|
|
prior assistant turns to round-trip on subsequent requests. The generic
|
|
third-party path strips them (signatures are Anthropic-proprietary and other
|
|
proxies cannot validate them), so without a DeepSeek-specific carve-out the
|
|
next tool-call turn fails with HTTP 400::
|
|
|
|
The content[].thinking in the thinking mode must be passed back to the
|
|
API.
|
|
|
|
DeepSeek's compatibility matrix lists ``thinking`` as supported but
|
|
``redacted_thinking`` and ``cache_control`` on thinking blocks as not
|
|
supported. Handling is the same as Kimi's ``/coding`` endpoint: strip
|
|
Anthropic-signed blocks (DeepSeek can't validate them) but preserve unsigned
|
|
blocks that Hermes synthesises from ``reasoning_content``.
|
|
|
|
See hermes-agent#16748.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import pytest
|
|
|
|
|
|
class TestDeepSeekAnthropicPreservesThinking:
|
|
"""convert_messages_to_anthropic must replay DeepSeek thinking blocks."""
|
|
|
|
|
|
|
|
def test_signed_anthropic_thinking_block_is_stripped(self) -> None:
|
|
"""Anthropic-signed blocks (that leaked through) must still be stripped.
|
|
|
|
DeepSeek issues its own signatures and cannot validate Anthropic's —
|
|
the strip-signed / keep-unsigned split matches the Kimi policy.
|
|
"""
|
|
from agent.anthropic_message_convert import convert_messages_to_anthropic
|
|
|
|
messages = [
|
|
{"role": "user", "content": "hi"},
|
|
{
|
|
"role": "assistant",
|
|
"content": [
|
|
{
|
|
"type": "thinking",
|
|
"thinking": "anthropic-signed payload",
|
|
"signature": "anthropic-sig-xyz",
|
|
},
|
|
{"type": "text", "text": "hello"},
|
|
],
|
|
},
|
|
{"role": "user", "content": "again"},
|
|
]
|
|
_system, converted = convert_messages_to_anthropic(
|
|
messages, base_url="https://api.deepseek.com/anthropic"
|
|
)
|
|
|
|
assistant_msg = next(m for m in converted if m["role"] == "assistant")
|
|
thinking_blocks = [
|
|
b for b in assistant_msg["content"]
|
|
if isinstance(b, dict) and b.get("type") == "thinking"
|
|
]
|
|
assert thinking_blocks == [], (
|
|
"Signed Anthropic thinking blocks must be stripped on DeepSeek — "
|
|
"DeepSeek cannot validate Anthropic-proprietary signatures."
|
|
)
|
|
|
|
def test_cache_control_stripped_from_thinking_block(self) -> None:
|
|
"""cache_control must still be stripped even when the block is preserved.
|
|
|
|
DeepSeek's compatibility matrix lists cache_control on thinking blocks
|
|
as ignored — cache markers interfere with signature validation on
|
|
upstreams that do check them, so Hermes strips them everywhere.
|
|
"""
|
|
from agent.anthropic_message_convert import convert_messages_to_anthropic
|
|
|
|
messages = [
|
|
{"role": "user", "content": "hi"},
|
|
{
|
|
"role": "assistant",
|
|
"reasoning_content": "r1",
|
|
"tool_calls": [
|
|
{
|
|
"id": "call_1",
|
|
"type": "function",
|
|
"function": {"name": "f", "arguments": "{}"},
|
|
}
|
|
],
|
|
},
|
|
{"role": "tool", "tool_call_id": "call_1", "content": "ok"},
|
|
]
|
|
# Inject cache_control on the synthesised thinking block after-the-fact
|
|
# by running conversion once, mutating, then re-running would be
|
|
# indirect. Instead check the simpler invariant: no thinking block in
|
|
# the converted output carries cache_control.
|
|
_system, converted = convert_messages_to_anthropic(
|
|
messages, base_url="https://api.deepseek.com/anthropic"
|
|
)
|
|
for m in converted:
|
|
if not isinstance(m.get("content"), list):
|
|
continue
|
|
for b in m["content"]:
|
|
if isinstance(b, dict) and b.get("type") in {"thinking", "redacted_thinking"}:
|
|
assert "cache_control" not in b
|
|
|
|
|
|
@pytest.mark.parametrize("model", [
|
|
"deepseek-r1", "deepseek-v4", "vendor/deepseek-v4-pro",
|
|
"gateway/deepseek-ai/deepseek_flash", " DeepSeek-Pro ", "deepseek_v4_flash",
|
|
])
|
|
def test_thinking_family_name_detection(model):
|
|
from agent.anthropic_endpoints import _model_name_is_deepseek_thinking
|
|
assert _model_name_is_deepseek_thinking(model)
|
|
|
|
|
|
@pytest.mark.parametrize("model", [None, "", " ", 42, "deepseek-chat", "deepseek-v3", "vendor/", "not-deepseek-v4", "qwen-thinking"])
|
|
def test_thinking_family_name_detection_rejects_unknown_models(model):
|
|
from agent.anthropic_endpoints import _model_name_is_deepseek_thinking
|
|
assert not _model_name_is_deepseek_thinking(model)
|
|
|
|
|
|
@pytest.mark.parametrize("url", [None, "https://api.anthropic.com", "https://inference-api.nousresearch.com/anthropic"])
|
|
def test_deepseek_model_name_does_not_override_native_signature_contract(url):
|
|
from agent.anthropic_message_convert import _manage_thinking_signatures
|
|
block = {"type": "thinking", "thinking": "signed native reasoning", "signature": "sig"}
|
|
messages = [{"role": "assistant", "content": [dict(block), {"type": "text", "text": "answer"}]}]
|
|
_manage_thinking_signatures(messages, url, "deepseek-v4")
|
|
assert messages[0]["content"][0] == block
|
|
|
|
|
|
def test_deepseek_proxy_keeps_unsigned_thinking_in_older_tool_turns_only():
|
|
import copy
|
|
from agent.anthropic_message_convert import convert_messages_to_anthropic
|
|
history = [
|
|
{"role": "user", "content": "inspect"},
|
|
{"role": "assistant", "content": "checking", "reasoning_details": [
|
|
{"type": "thinking", "thinking": "unsigned", "cache_control": {"type": "ephemeral"}},
|
|
{"type": "thinking", "thinking": "foreign signed", "signature": "sig"},
|
|
{"type": "redacted_thinking", "data": "redacted-signature"},
|
|
], "tool_calls": [{"id": "call_1", "type": "function", "function": {"name": "inspect", "arguments": "{}"}}]},
|
|
{"role": "tool", "tool_call_id": "call_1", "content": "ok"},
|
|
{"role": "assistant", "content": "done"},
|
|
]
|
|
snapshot = copy.deepcopy(history)
|
|
_, result = convert_messages_to_anthropic(history, base_url="https://proxy.example/anthropic", model="vendor/deepseek-v4")
|
|
assistant = next(m for m in result if m["role"] == "assistant")
|
|
assert [b for b in assistant["content"] if b.get("type") in {"thinking", "redacted_thinking"}] == [
|
|
{"type": "thinking", "thinking": "unsigned"},
|
|
]
|
|
assert history == snapshot
|
|
|