mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-05-09 03:11:58 +00:00
812ce0b987
When empty-response terminal scaffolding fires on a tool-result turn, _drop_trailing_empty_response_scaffolding left the live history ending at a bare 'tool' message. The next user input then landed as [...tool, user], a protocol-invalid sequence that OpenRouter/Opus and other providers silently fail on (returns empty content). That retriggered the empty-retry recovery every turn, and recovery flags never hit SQLite (no column for them), so history kept looking broken on every reload. Two fixes: 1. Scaffolding strip rewinds the orphan assistant(tool_calls)+tool pair after popping sentinels. Only fires when scaffolding flags were actually present, so mid-iteration tool loops are untouched. 2. _repair_message_sequence runs right before every API call as a defensive belt: drops stray tool messages with unknown tool_call_ids, merges consecutive user messages so no user input is lost. Does NOT rewind assistant(tool_calls)+tool+user — that pattern is valid when the user redirected before the model got its continuation turn. Repro: session 20260507_044111_fa7e65. Opus-4.7/OpenRouter returned content-less response after a 42KB execute_code output, nudge+retry chain exhausted (no fallback configured), terminal sentinel appended, scaffolding stripped leaving bare tool tail, user typed 'wtf happened..' and landed as tool→user violation. Every subsequent turn collapsed in <50ms with the same 3-retry empty chain because the API request itself was malformed. Verified live via HTTP mock: pre-fix reproduced 5 api_calls/0.15s exit 'empty_response_exhausted'; post-fix 1 api_call/0.10s exit 'text_response(finish_reason=stop)'. Three-turn session flows cleanly through the scenario. Full run_agent suite: 1242 passed (0 regressions, 2 pre-existing concurrent_interrupt failures unrelated).
98 lines
3.5 KiB
Python
98 lines
3.5 KiB
Python
"""Regression tests for empty-response recovery transcript persistence."""
|
|
|
|
from run_agent import AIAgent
|
|
|
|
|
|
def _agent_with_stubbed_persistence():
|
|
agent = AIAgent.__new__(AIAgent)
|
|
agent._persist_user_message_idx = None
|
|
agent._persist_user_message_override = None
|
|
agent._session_db = None
|
|
agent._session_messages = []
|
|
agent.saved_session_logs = []
|
|
agent.flushed_session_db_messages = []
|
|
agent._save_session_log = lambda messages: agent.saved_session_logs.append(
|
|
[m.copy() for m in messages]
|
|
)
|
|
agent._flush_messages_to_session_db = lambda messages, conversation_history=None: (
|
|
agent.flushed_session_db_messages.append([m.copy() for m in messages])
|
|
)
|
|
return agent
|
|
|
|
|
|
def test_persist_session_strips_trailing_empty_recovery_scaffolding():
|
|
"""After stripping scaffolding, also rewind past orphan trailing tool-result
|
|
messages that the failed iteration left behind. Otherwise the next user
|
|
message lands after a bare ``tool`` and produces a protocol-invalid
|
|
sequence that most providers silently fail on, retriggering the empty-
|
|
retry loop indefinitely.
|
|
"""
|
|
agent = _agent_with_stubbed_persistence()
|
|
messages = [
|
|
{"role": "user", "content": "run the task"},
|
|
{
|
|
"role": "assistant",
|
|
"content": "",
|
|
"tool_calls": [{"id": "call_1", "type": "function",
|
|
"function": {"name": "x", "arguments": "{}"}}],
|
|
},
|
|
{"role": "tool", "content": "{}", "tool_call_id": "call_1"},
|
|
{
|
|
"role": "assistant",
|
|
"content": "(empty)",
|
|
"_empty_recovery_synthetic": True,
|
|
},
|
|
{
|
|
"role": "user",
|
|
"content": (
|
|
"You just executed tool calls but returned an empty response. "
|
|
"Please process the tool results above and continue with the task."
|
|
),
|
|
"_empty_recovery_synthetic": True,
|
|
},
|
|
]
|
|
|
|
AIAgent._persist_session(agent, messages, conversation_history=[])
|
|
|
|
# After strip + rewind, only the original user message remains. The
|
|
# assistant(tool_calls) + tool pair is dropped because its iteration
|
|
# never produced a real response.
|
|
assert messages == [
|
|
{"role": "user", "content": "run the task"},
|
|
]
|
|
assert agent.saved_session_logs[-1] == messages
|
|
assert all(not msg.get("_empty_recovery_synthetic") for msg in messages)
|
|
|
|
|
|
def test_persist_session_keeps_unmarked_terminal_empty_response():
|
|
agent = _agent_with_stubbed_persistence()
|
|
messages = [
|
|
{"role": "user", "content": "run the task"},
|
|
{"role": "assistant", "content": "(empty)"},
|
|
]
|
|
|
|
AIAgent._persist_session(agent, messages, conversation_history=[])
|
|
|
|
assert messages == [
|
|
{"role": "user", "content": "run the task"},
|
|
{"role": "assistant", "content": "(empty)"},
|
|
]
|
|
assert agent.saved_session_logs[-1] == messages
|
|
|
|
|
|
def test_persist_session_strips_marked_terminal_empty_sentinel():
|
|
agent = _agent_with_stubbed_persistence()
|
|
messages = [
|
|
{"role": "user", "content": "continue"},
|
|
{
|
|
"role": "assistant",
|
|
"content": "(empty)",
|
|
"_empty_terminal_sentinel": True,
|
|
},
|
|
]
|
|
|
|
AIAgent._persist_session(agent, messages, conversation_history=[])
|
|
|
|
assert messages == [{"role": "user", "content": "continue"}]
|
|
assert agent.saved_session_logs[-1] == messages
|
|
assert all(not msg.get("_empty_terminal_sentinel") for msg in messages)
|