mirrors/hermes-agent
1
0
Fork 0
mirror of https://github.com/NousResearch/hermes-agent.git synced 2026-07-17 14:42:06 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 6

feat(status): restore model and context in gateway status

Browse source 
PROBLEM: The old public /status PR drifted out of the current Amy patch stack, leaving /status without the model/provider, context window, or explicit cumulative token label that Wolfram uses to monitor context pressure from chat.

SOLUTION: Re-port the feature onto the current gateway status handler. Prefer live/cached agent runtime metadata, fall back to SessionDB + SessionStore state between turns, add localized status model/context lines, and keep token totals explicitly labeled cumulative.

Verification: tests/gateway/test_status_command.py, tests/hermes_cli/test_commands.py
This commit is contained in:
Wolfram Ravenwolf 2026-05-31 17:14:47 +02:00 • committed by Teknium
parent 5035fa9029
commit ead38107a2
19 changed files with 257 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

73
tests/gateway/test_status_command.py
View file

@ -174,6 +174,79 @@ async def test_status_command_tokens_zero_when_session_db_row_missing():
assert "**Cumulative API tokens (re-sent each call):** 0" in result
@pytest.mark.asyncio
async def test_status_command_includes_live_agent_model_and_context():
session_entry = SessionEntry(
session_key=build_session_key(_make_source()),
session_id="sess-1",
created_at=datetime.now(),
updated_at=datetime.now(),
platform=Platform.TELEGRAM,
chat_type="dm",
total_tokens=0,
)
runner = _make_runner(session_entry)
runner._session_db.get_session.return_value = {
"input_tokens": 1000,
"output_tokens": 250,
"cache_read_tokens": 0,
"cache_write_tokens": 0,
"reasoning_tokens": 0,
"model": "openai/gpt-test",
}
running_agent = SimpleNamespace(
model="openai/gpt-test",
provider="openai",
context_compressor=SimpleNamespace(
last_prompt_tokens=12_345,
context_length=100_000,
),
interrupt=MagicMock(),
)
runner._running_agents[build_session_key(_make_source())] = running_agent
result = await runner._handle_message(_make_event("/status"))
assert "**Model:** `openai/gpt-test` (openai)" in result
assert "**Context:** 12,345 / 100,000 (12%)" in result
assert "**Cumulative API tokens (re-sent each call):** 1,250 (cumulative)" in result
@pytest.mark.asyncio
async def test_status_command_includes_persisted_model_and_context_when_agent_not_running(monkeypatch):
session_entry = SessionEntry(
session_key=build_session_key(_make_source()),
session_id="sess-1",
created_at=datetime.now(),
updated_at=datetime.now(),
platform=Platform.TELEGRAM,
chat_type="dm",
total_tokens=0,
last_prompt_tokens=24_000,
)
runner = _make_runner(session_entry)
runner._session_db.get_session.return_value = {
"input_tokens": 2000,
"output_tokens": 500,
"cache_read_tokens": 0,
"cache_write_tokens": 0,
"reasoning_tokens": 0,
"model": "openai/gpt-persisted",
"billing_provider": "openai-codex",
"billing_base_url": "https://example.invalid/v1",
}
monkeypatch.setattr(
"agent.model_metadata.get_model_context_length",
lambda *_args, **_kwargs: 272_000,
)
result = await runner._handle_message(_make_event("/status"))
assert "**Model:** `openai/gpt-persisted` (openai-codex)" in result
assert "**Context:** 24,000 / 272,000 (9%)" in result
assert "**Cumulative API tokens (re-sent each call):** 2,500 (cumulative)" in result
@pytest.mark.asyncio
async def test_agents_command_reports_active_agents_and_processes(monkeypatch):
session_key = build_session_key(_make_source())