seb/hermes-agent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'main' of github.com:NousResearch/hermes-agent into bb/gui

Browse Source
This commit is contained in:
Brooklyn Nicholson
2026-05-04 12:47:53 -05:00
parent 27c5fa5381 0ce1b9fe20
commit ca8f2c7907
182 changed files with 9843 additions and 974 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/agent/test_anthropic_adapter.py
+39
View File
@@ -1113,6 +1113,45 @@ class TestBuildAnthropicKwargs:
assert _forbids_sampling_params("claude-opus-4-6") is False
assert _forbids_sampling_params("claude-sonnet-4-5") is False
def test_supports_fast_mode_predicate(self):
"""Fast mode is Opus 4.6 only — Opus 4.7 and others must be excluded."""
from agent.anthropic_adapter import _supports_fast_mode
assert _supports_fast_mode("claude-opus-4-6") is True
assert _supports_fast_mode("anthropic/claude-opus-4-6") is True
assert _supports_fast_mode("claude-opus-4-7") is False
assert _supports_fast_mode("claude-sonnet-4-6") is False
assert _supports_fast_mode("claude-haiku-4-5") is False
assert _supports_fast_mode("") is False
def test_fast_mode_omitted_for_unsupported_model(self):
"""fast_mode=True on Opus 4.7 must NOT inject speed=fast (API 400s)."""
kwargs = build_anthropic_kwargs(
model="claude-opus-4-7",
messages=[{"role": "user", "content": "hi"}],
tools=None,
max_tokens=1024,
reasoning_config=None,
fast_mode=True,
)
# extra_body either absent or doesn't carry "speed"
assert "speed" not in kwargs.get("extra_body", {})
# No fast-mode beta header should be added either
beta_header = (kwargs.get("extra_headers") or {}).get("anthropic-beta", "")
assert "fast-mode-2026-02-01" not in beta_header
def test_fast_mode_still_applied_on_opus_46(self):
"""Regression guard — fast mode must still work on Opus 4.6."""
kwargs = build_anthropic_kwargs(
model="claude-opus-4-6",
messages=[{"role": "user", "content": "hi"}],
tools=None,
max_tokens=1024,
reasoning_config=None,
fast_mode=True,
)
assert kwargs.get("extra_body", {}).get("speed") == "fast"
assert "fast-mode-2026-02-01" in kwargs["extra_headers"]["anthropic-beta"]
def test_reasoning_disabled(self):
kwargs = build_anthropic_kwargs(
model="claude-sonnet-4-20250514",
tests/agent/test_auxiliary_client.py
+50
View File
@@ -1893,3 +1893,53 @@ class TestOpenRouterExplicitApiKey:
assert call_kwargs["api_key"] == "env-fallback-key", (
f"Expected env fallback key to be used when explicit_api_key is None, got: {call_kwargs['api_key']}"
)
class TestAnthropicExplicitApiKey:
"""Test that explicit_api_key is correctly propagated to _try_anthropic().
Parity with the OpenRouter fix in #18768: resolve_provider_client() passes
explicit_api_key to _try_openrouter(), but the anthropic branch was not
updated — _try_anthropic() always fell back to resolve_anthropic_token()
even when an explicit key was supplied (e.g. from a fallback_model entry).
"""
def test_try_anthropic_uses_explicit_api_key_over_env(self):
"""_try_anthropic(explicit_api_key) must use the supplied key, not the env fallback."""
with patch("agent.anthropic_adapter.resolve_anthropic_token", return_value="env-fallback-key"), \
patch("agent.anthropic_adapter.build_anthropic_client") as mock_build, \
patch("agent.auxiliary_client._select_pool_entry", return_value=(False, None)):
mock_build.return_value = MagicMock()
from agent.auxiliary_client import _try_anthropic
client, model = _try_anthropic("explicit-pool-key")
assert client is not None
assert mock_build.call_args.args[0] == "explicit-pool-key", (
f"Expected explicit_api_key to be passed, got: {mock_build.call_args.args[0]}"
)
assert mock_build.call_args.args[0] != "env-fallback-key"
def test_try_anthropic_without_explicit_key_falls_back_to_resolve(self):
"""Without explicit_api_key, _try_anthropic falls back to resolve_anthropic_token."""
with patch("agent.anthropic_adapter.resolve_anthropic_token", return_value="env-fallback-key"), \
patch("agent.anthropic_adapter.build_anthropic_client") as mock_build, \
patch("agent.auxiliary_client._select_pool_entry", return_value=(False, None)):
mock_build.return_value = MagicMock()
from agent.auxiliary_client import _try_anthropic
client, model = _try_anthropic()
assert client is not None
assert mock_build.call_args.args[0] == "env-fallback-key"
def test_resolve_provider_client_passes_explicit_api_key_to_anthropic(self):
"""resolve_provider_client(provider='anthropic', explicit_api_key=...) must propagate the key."""
with patch("agent.anthropic_adapter.resolve_anthropic_token", return_value="env-key"), \
patch("agent.anthropic_adapter.build_anthropic_client") as mock_build, \
patch("agent.auxiliary_client._select_pool_entry", return_value=(False, None)):
mock_build.return_value = MagicMock()
client, model = resolve_provider_client(
provider="anthropic",
explicit_api_key="explicit-fallback-key",
)
assert client is not None
assert mock_build.call_args.args[0] == "explicit-fallback-key", (
"resolve_provider_client must forward explicit_api_key to _try_anthropic()"
)
tests/agent/test_bedrock_adapter.py
+6
View File
@@ -1283,18 +1283,21 @@ class TestIsStaleConnectionError:
"""Classifier that decides whether an exception warrants client eviction."""
def test_detects_botocore_connection_closed_error(self):
pytest.importorskip("botocore", reason="botocore required for Bedrock exception tests")
from agent.bedrock_adapter import is_stale_connection_error
from botocore.exceptions import ConnectionClosedError
exc = ConnectionClosedError(endpoint_url="https://bedrock.example")
assert is_stale_connection_error(exc) is True
def test_detects_botocore_endpoint_connection_error(self):
pytest.importorskip("botocore", reason="botocore required for Bedrock exception tests")
from agent.bedrock_adapter import is_stale_connection_error
from botocore.exceptions import EndpointConnectionError
exc = EndpointConnectionError(endpoint_url="https://bedrock.example")
assert is_stale_connection_error(exc) is True
def test_detects_botocore_read_timeout(self):
pytest.importorskip("botocore", reason="botocore required for Bedrock exception tests")
from agent.bedrock_adapter import is_stale_connection_error
from botocore.exceptions import ReadTimeoutError
exc = ReadTimeoutError(endpoint_url="https://bedrock.example")
@@ -1355,6 +1358,7 @@ class TestCallConverseInvalidatesOnStaleError:
reconnects instead of reusing the dead socket."""
def test_converse_evicts_client_on_stale_error(self):
pytest.importorskip("botocore", reason="botocore required for Bedrock exception tests")
from agent.bedrock_adapter import (
_bedrock_runtime_client_cache,
call_converse,
@@ -1381,6 +1385,7 @@ class TestCallConverseInvalidatesOnStaleError:
)
def test_converse_stream_evicts_client_on_stale_error(self):
pytest.importorskip("botocore", reason="botocore required for Bedrock exception tests")
from agent.bedrock_adapter import (
_bedrock_runtime_client_cache,
call_converse_stream,
@@ -1406,6 +1411,7 @@ class TestCallConverseInvalidatesOnStaleError:
def test_converse_does_not_evict_on_non_stale_error(self):
"""Non-stale errors (e.g. ValidationException) leave the client cache alone."""
pytest.importorskip("botocore", reason="botocore required for Bedrock exception tests")
from agent.bedrock_adapter import (
_bedrock_runtime_client_cache,
call_converse,
tests/agent/test_context_compressor.py
+41
View File
@@ -1281,6 +1281,47 @@ class TestTokenBudgetTailProtection:
assert isinstance(cut, int)
assert 0 <= cut <= len(messages)
def test_generous_budget_protects_everything_floor_does_not_override(
self, budget_compressor
):
"""A budget that covers the whole transcript must prune nothing —
``protect_tail_count`` is a minimum floor, not a ceiling."""
c = budget_compressor
# 100 alternating assistant/tool messages. Each tool result has
# *unique* content so the dedup pass (Pass 1, which is independent
# of prune_boundary) is a no-op and we isolate the boundary logic.
messages = []
for i in range(50):
messages.append({
"role": "assistant", "content": None,
"tool_calls": [{
"id": f"c{i}",
"type": "function",
"function": {"name": "noop", "arguments": "{}"},
}],
})
messages.append({
"role": "tool",
"tool_call_id": f"c{i}",
"content": f"unique-tool-output-{i:03d}-" + ("x" * 250),
})
# Budget large enough to cover the whole transcript many times over,
# so the budget walk completes without hitting its break condition
# and the boundary lands at 0 ("protect everything").
_, pruned = c._prune_old_tool_results(
messages,
protect_tail_count=20,
protect_tail_tokens=10_000_000,
)
assert pruned == 0, (
"budget said protect everything, but the floor still pruned "
f"{pruned} messages — protect_tail_count is acting as a ceiling, "
"not a minimum floor"
)
class TestUpdateModelBudgets:
"""Regression: update_model() must recalculate token budgets."""
tests/agent/test_curator.py
+117
View File
@@ -154,6 +154,7 @@ def test_unused_skill_transitions_to_stale(curator_env):
long_ago = (datetime.now(timezone.utc) - timedelta(days=45)).isoformat()
data = u.load_usage()
data["old-skill"] = u._empty_record()
data["old-skill"]["created_by"] = "agent"
data["old-skill"]["last_used_at"] = long_ago
data["old-skill"]["created_at"] = long_ago
u.save_usage(data)
@@ -172,6 +173,7 @@ def test_very_old_skill_gets_archived(curator_env):
super_old = (datetime.now(timezone.utc) - timedelta(days=120)).isoformat()
data = u.load_usage()
data["ancient"] = u._empty_record()
data["ancient"]["created_by"] = "agent"
data["ancient"]["last_used_at"] = super_old
data["ancient"]["created_at"] = super_old
u.save_usage(data)
@@ -192,6 +194,7 @@ def test_pinned_skill_is_never_touched(curator_env):
super_old = (datetime.now(timezone.utc) - timedelta(days=365)).isoformat()
data = u.load_usage()
data["precious"] = u._empty_record()
data["precious"]["created_by"] = "agent"
data["precious"]["last_used_at"] = super_old
data["precious"]["created_at"] = super_old
data["precious"]["pinned"] = True
@@ -214,6 +217,7 @@ def test_stale_skill_reactivates_on_recent_use(curator_env):
recent = datetime.now(timezone.utc).isoformat()
data = u.load_usage()
data["revived"] = u._empty_record()
data["revived"]["created_by"] = "agent"
data["revived"]["state"] = "stale"
data["revived"]["last_used_at"] = recent
data["revived"]["created_at"] = recent
@@ -240,6 +244,27 @@ def test_new_skill_without_last_used_not_immediately_archived(curator_env):
assert (skills_dir / "fresh").exists()
def test_manual_skill_is_not_auto_archived(curator_env):
"""Manual skills can have usage records, but without the agent-created
marker they must stay out of curator transitions."""
c = curator_env["curator"]
u = curator_env["usage"]
skills_dir = curator_env["home"] / "skills"
skill_dir = _write_skill(skills_dir, "manual")
super_old = (datetime.now(timezone.utc) - timedelta(days=365)).isoformat()
data = u.load_usage()
data["manual"] = u._empty_record()
data["manual"]["last_used_at"] = super_old
data["manual"]["created_at"] = super_old
u.save_usage(data)
counts = c.apply_automatic_transitions()
assert counts["checked"] == 0
assert counts["archived"] == 0
assert skill_dir.exists()
def test_bundled_skill_not_touched_by_transitions(curator_env):
c = curator_env["curator"]
u = curator_env["usage"]
@@ -267,8 +292,10 @@ def test_bundled_skill_not_touched_by_transitions(curator_env):
def test_run_review_records_state(curator_env):
c = curator_env["curator"]
u = curator_env["usage"]
skills_dir = curator_env["home"] / "skills"
_write_skill(skills_dir, "a")
u.mark_agent_created("a")
result = c.run_curator_review(synchronous=True)
assert "started_at" in result
@@ -284,8 +311,10 @@ def test_dry_run_does_not_advance_state(curator_env, monkeypatch):
`hermes curator status`. Fixes #18373.
"""
c = curator_env["curator"]
u = curator_env["usage"]
skills_dir = curator_env["home"] / "skills"
_write_skill(skills_dir, "a")
u.mark_agent_created("a")
# Stub the LLM so the test doesn't need a provider.
monkeypatch.setattr(
@@ -311,8 +340,10 @@ def test_dry_run_injects_report_only_banner(curator_env, monkeypatch):
skips automatic transitions — but the LLM prompt is the only guard
against the model calling skill_manage directly."""
c = curator_env["curator"]
u = curator_env["usage"]
skills_dir = curator_env["home"] / "skills"
_write_skill(skills_dir, "a")
u.mark_agent_created("a")
captured = {}
def _stub(prompt):
@@ -331,8 +362,10 @@ def test_dry_run_skips_automatic_transitions(curator_env, monkeypatch):
archives skills deterministically, and a preview must not touch the
filesystem."""
c = curator_env["curator"]
u = curator_env["usage"]
skills_dir = curator_env["home"] / "skills"
_write_skill(skills_dir, "a")
u.mark_agent_created("a")
called = {"n": 0}
def _explode(*_a, **_kw):
@@ -351,8 +384,10 @@ def test_dry_run_skips_automatic_transitions(curator_env, monkeypatch):
def test_run_review_synchronous_invokes_llm_stub(curator_env, monkeypatch):
c = curator_env["curator"]
u = curator_env["usage"]
skills_dir = curator_env["home"] / "skills"
_write_skill(skills_dir, "a")
u.mark_agent_created("a")
calls = []
def _stub(prompt):
@@ -409,8 +444,10 @@ def test_maybe_run_curator_enforces_idle_gate(curator_env, monkeypatch):
def test_maybe_run_curator_runs_when_eligible(curator_env, monkeypatch):
c = curator_env["curator"]
u = curator_env["usage"]
skills_dir = curator_env["home"] / "skills"
_write_skill(skills_dir, "a")
u.mark_agent_created("a")
# Seed last_run_at far in the past so the interval gate opens — the
# "no state" path intentionally defers the first run now (#18373).
long_ago = datetime.now(timezone.utc) - timedelta(hours=c.get_interval_hours() * 2)
@@ -645,6 +682,86 @@ def test_review_model_honors_auxiliary_curator_slot(curator_env):
)
def test_review_runtime_passes_auxiliary_curator_credentials(curator_env):
"""Per-slot api_key/base_url must ride into resolve_runtime_provider (not main-only creds)."""
curator = curator_env["curator"]
cfg = {
"model": {"provider": "openrouter", "default": "openai/gpt-5.5"},
"auxiliary": {
"curator": {
"provider": "custom",
"model": "local-mini",
"api_key": "sk-curator-only",
"base_url": "http://localhost:11434/v1",
},
},
}
binding = curator._resolve_review_runtime(cfg)
assert binding.provider == "custom"
assert binding.model == "local-mini"
assert binding.explicit_api_key == "sk-curator-only"
assert binding.explicit_base_url == "http://localhost:11434/v1"
def test_review_runtime_strips_blank_aux_credentials(curator_env):
curator = curator_env["curator"]
cfg = {
"model": {"provider": "openrouter", "default": "openai/gpt-5.5"},
"auxiliary": {
"curator": {
"provider": "openrouter",
"model": "x/y",
"api_key": " ",
"base_url": "",
},
},
}
binding = curator._resolve_review_runtime(cfg)
assert binding.explicit_api_key is None
assert binding.explicit_base_url is None
def test_review_runtime_ignores_auxiliary_credentials_when_using_main(curator_env):
"""Falling through to main model must not pick up stray auxiliary.curator secrets."""
curator = curator_env["curator"]
cfg = {
"model": {"provider": "openrouter", "default": "openai/gpt-5.5"},
"auxiliary": {
"curator": {
"provider": "auto",
"model": "",
"api_key": "must-not-leak",
"base_url": "http://curator-slot-ignored/",
},
},
}
binding = curator._resolve_review_runtime(cfg)
assert (binding.provider, binding.model) == ("openrouter", "openai/gpt-5.5")
assert binding.explicit_api_key is None
assert binding.explicit_base_url is None
def test_review_runtime_legacy_auxiliary_carry_credentials(curator_env, caplog):
curator = curator_env["curator"]
cfg = {
"model": {"provider": "openrouter", "default": "openai/gpt-5.5"},
"curator": {
"auxiliary": {
"provider": "custom",
"model": "m",
"api_key": "legacy-key",
"base_url": "http://legacy/v1",
},
},
}
import logging
with caplog.at_level(logging.INFO, logger="agent.curator"):
binding = curator._resolve_review_runtime(cfg)
assert binding.explicit_api_key == "legacy-key"
assert binding.explicit_base_url == "http://legacy/v1"
assert any("deprecated curator.auxiliary" in rec.message for rec in caplog.records)
def test_review_model_auxiliary_curator_partial_override_falls_back(curator_env):
"""Only one of slot provider/model set → fall back to the main pair.
tests/agent/test_curator_classification.py
+75
View File
@@ -220,6 +220,81 @@ def test_classify_handles_malformed_arguments_string(curator_env):
assert len(result["pruned"]) == 1
def test_classify_no_false_positive_short_name_in_file_path(curator_env):
"""Short skill name that is a substring of another filename = pruned, not consolidated."""
# e.g. "api" should NOT match "references/api-design.md"
result = curator_env._classify_removed_skills(
removed=["api"],
added=[],
after_names={"conventions"},
tool_calls=[
{
"name": "skill_manage",
"arguments": json.dumps({
"action": "write_file",
"name": "conventions",
"file_path": "references/api-design.md",
"file_content": "# API Design\n...",
}),
},
],
)
assert result["consolidated"] == [], (
f"Short name 'api' should NOT match file_path 'references/api-design.md'"
)
assert len(result["pruned"]) == 1
assert result["pruned"][0]["name"] == "api"
def test_classify_no_false_positive_short_name_in_content(curator_env):
"""Short skill name embedded in longer word in content = pruned, not consolidated."""
# e.g. "test" should NOT match content "running latest tests"
result = curator_env._classify_removed_skills(
removed=["test"],
added=[],
after_names={"umbrella"},
tool_calls=[
{
"name": "skill_manage",
"arguments": json.dumps({
"action": "patch",
"name": "umbrella",
"old_string": "old",
"new_string": "running latest tests with pytest",
}),
},
],
)
assert result["consolidated"] == [], (
f"Short name 'test' should NOT match 'latest' via word boundary"
)
assert len(result["pruned"]) == 1
def test_classify_still_matches_exact_word_in_content(curator_env):
"""Word-boundary match still works for exact word occurrences."""
# "api" SHOULD match content "use the api gateway"
result = curator_env._classify_removed_skills(
removed=["api"],
added=[],
after_names={"gateway"},
tool_calls=[
{
"name": "skill_manage",
"arguments": json.dumps({
"action": "edit",
"name": "gateway",
"content": "# Gateway\n\nUse the api gateway for all requests.\n",
}),
},
],
)
assert len(result["consolidated"]) == 1, (
f"'api' should match as a standalone word in content"
)
assert result["consolidated"][0]["into"] == "gateway"
def test_report_md_splits_consolidated_and_pruned_sections(curator_env):
"""End-to-end: REPORT.md shows both sections distinctly."""
curator = curator_env
tests/agent/test_error_classifier.py
+32
View File
@@ -410,6 +410,24 @@ class TestClassifyApiError:
result = classify_api_error(e, approx_tokens=1000, context_length=200000)
assert result.reason == FailoverReason.format_error
def test_400_generic_many_messages_below_large_context_pressure_is_format_error(self):
"""Large-context sessions should not overflow solely due to message count."""
e = MockAPIError(
"Error",
status_code=400,
body={"error": {"message": "Error"}},
)
result = classify_api_error(
e,
provider="openai-codex",
model="gpt-5.5",
approx_tokens=74320,
context_length=1_000_000,
num_messages=432,
)
assert result.reason == FailoverReason.format_error
assert result.should_compress is False
# ── Server disconnect + large session ──
def test_disconnect_large_session_context_overflow(self):
@@ -425,6 +443,20 @@ class TestClassifyApiError:
result = classify_api_error(e, approx_tokens=5000, context_length=200000)
assert result.reason == FailoverReason.timeout
def test_disconnect_many_messages_below_large_context_pressure_is_timeout(self):
"""Large-context disconnects should not overflow solely due to message count."""
e = Exception("server disconnected without sending complete message")
result = classify_api_error(
e,
provider="openai-codex",
model="gpt-5.5",
approx_tokens=74320,
context_length=1_000_000,
num_messages=432,
)
assert result.reason == FailoverReason.timeout
assert result.should_compress is False
# ── Provider-specific: Anthropic thinking signature ──
def test_anthropic_thinking_signature(self):
tests/agent/test_vision_resolved_args.py
+34 -10
View File
@@ -13,16 +13,13 @@ def test_vision_call_uses_resolved_provider_args():
usage=MagicMock(prompt_tokens=10, completion_tokens=5),
)
with (
patch(
"agent.auxiliary_client._resolve_task_provider_model",
return_value=("my-resolved-provider", "my-resolved-model", "http://resolved", "resolved-key", "chat_completions"),
),
patch(
"agent.auxiliary_client.resolve_vision_provider_client",
return_value=("my-resolved-provider", fake_client, "my-resolved-model"),
) as mock_vision,
):
with patch(
"agent.auxiliary_client._resolve_task_provider_model",
return_value=("my-resolved-provider", "my-resolved-model", "http://resolved", "resolved-key", "chat_completions"),
), patch(
"agent.auxiliary_client.resolve_vision_provider_client",
return_value=("my-resolved-provider", fake_client, "my-resolved-model"),
) as mock_vision:
call_llm(
"vision",
provider="raw-provider",
@@ -38,3 +35,30 @@ def test_vision_call_uses_resolved_provider_args():
assert call_args.kwargs["model"] == "my-resolved-model"
assert call_args.kwargs["base_url"] == "http://resolved"
assert call_args.kwargs["api_key"] == "resolved-key"
def test_vision_base_url_override_keeps_explicit_provider():
"""Explicit provider should still drive credential resolution with custom base_url."""
from agent.auxiliary_client import resolve_vision_provider_client
fake_client = MagicMock()
with patch(
"agent.auxiliary_client._resolve_task_provider_model",
return_value=(
"zai",
"glm-4v",
"https://open.bigmodel.cn/api/paas/v4",
None,
"chat_completions",
),
), patch(
"agent.auxiliary_client.resolve_provider_client",
return_value=(fake_client, "glm-4v"),
) as mock_resolve:
provider, client, model = resolve_vision_provider_client()
assert provider == "zai"
assert client is fake_client
assert model == "glm-4v"
assert mock_resolve.call_args.args[0] == "zai"
assert mock_resolve.call_args.kwargs["explicit_base_url"] == "https://open.bigmodel.cn/api/paas/v4"
tests/agent/transports/test_codex_transport.py
+14
View File
@@ -126,6 +126,20 @@ class TestCodexBuildKwargs:
)
assert kw.get("extra_headers", {}).get("x-grok-conv-id") == "conv-123"
def test_xai_headers_preserve_request_override_headers(self, transport):
messages = [{"role": "user", "content": "Hi"}]
kw = transport.build_kwargs(
model="grok-3", messages=messages, tools=[],
session_id="conv-123",
is_xai_responses=True,
request_overrides={"extra_headers": {"X-Test": "1", "X-Trace": "abc"}},
)
assert kw.get("extra_headers") == {
"X-Test": "1",
"X-Trace": "abc",
"x-grok-conv-id": "conv-123",
}
def test_minimal_effort_clamped(self, transport):
messages = [{"role": "user", "content": "Hi"}]
kw = transport.build_kwargs(