fix: remove legacy compression.summary_* config and env var fallbacks (#8992)

Remove the backward-compat code paths that read compression provider/model
settings from legacy config keys and env vars, which caused silent failures
when auto-detection resolved to incompatible backends.

What changed:
- Remove compression.summary_model, summary_provider, summary_base_url from
  DEFAULT_CONFIG and cli.py defaults
- Remove backward-compat block in _resolve_task_provider_model() that read
  from the legacy compression section
- Remove _get_auxiliary_provider() and _get_auxiliary_env_override() helper
  functions (AUXILIARY_*/CONTEXT_* env var readers)
- Remove env var fallback chain for per-task overrides
- Update hermes config show to read from auxiliary.compression
- Add config migration (v16→17) that moves non-empty legacy values to
  auxiliary.compression and strips the old keys
- Update example config and openclaw migration script
- Remove/update tests for deleted code paths

Compression model/provider is now configured exclusively via:
  auxiliary.compression.provider / auxiliary.compression.model

Closes #8923
This commit is contained in:
Teknium
2026-04-13 04:59:26 -07:00
committed by GitHub
parent c1809e85e7
commit e3ffe5b75f
12 changed files with 58 additions and 224 deletions

View File

@@ -17,7 +17,6 @@ from agent.auxiliary_client import (
call_llm,
async_call_llm,
_read_codex_access_token,
_get_auxiliary_provider,
_get_provider_chain,
_is_payment_error,
_try_payment_fallback,
@@ -32,12 +31,6 @@ def _clean_env(monkeypatch):
"OPENROUTER_API_KEY", "OPENAI_BASE_URL", "OPENAI_API_KEY",
"OPENAI_MODEL", "LLM_MODEL", "NOUS_INFERENCE_BASE_URL",
"ANTHROPIC_API_KEY", "ANTHROPIC_TOKEN", "CLAUDE_CODE_OAUTH_TOKEN",
# Per-task provider/model/direct-endpoint overrides
"AUXILIARY_VISION_PROVIDER", "AUXILIARY_VISION_MODEL",
"AUXILIARY_VISION_BASE_URL", "AUXILIARY_VISION_API_KEY",
"AUXILIARY_WEB_EXTRACT_PROVIDER", "AUXILIARY_WEB_EXTRACT_MODEL",
"AUXILIARY_WEB_EXTRACT_BASE_URL", "AUXILIARY_WEB_EXTRACT_API_KEY",
"CONTEXT_COMPRESSION_PROVIDER", "CONTEXT_COMPRESSION_MODEL",
):
monkeypatch.delenv(key, raising=False)
@@ -568,29 +561,6 @@ class TestGetTextAuxiliaryClient:
call_kwargs = mock_openai.call_args
assert call_kwargs.kwargs["base_url"] == "http://localhost:1234/v1"
def test_task_direct_endpoint_override(self, monkeypatch):
monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
monkeypatch.setenv("AUXILIARY_WEB_EXTRACT_BASE_URL", "http://localhost:2345/v1")
monkeypatch.setenv("AUXILIARY_WEB_EXTRACT_API_KEY", "task-key")
monkeypatch.setenv("AUXILIARY_WEB_EXTRACT_MODEL", "task-model")
with patch("agent.auxiliary_client.OpenAI") as mock_openai:
client, model = get_text_auxiliary_client("web_extract")
assert model == "task-model"
assert mock_openai.call_args.kwargs["base_url"] == "http://localhost:2345/v1"
assert mock_openai.call_args.kwargs["api_key"] == "task-key"
def test_task_direct_endpoint_without_openai_key_uses_placeholder(self, monkeypatch):
"""Local endpoints without an API key should use 'no-key-required' placeholder."""
monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
monkeypatch.setenv("AUXILIARY_WEB_EXTRACT_BASE_URL", "http://localhost:2345/v1")
monkeypatch.setenv("AUXILIARY_WEB_EXTRACT_MODEL", "task-model")
with patch("agent.auxiliary_client.OpenAI") as mock_openai:
client, model = get_text_auxiliary_client("web_extract")
assert client is not None
assert model == "task-model"
assert mock_openai.call_args.kwargs["api_key"] == "no-key-required"
assert mock_openai.call_args.kwargs["base_url"] == "http://localhost:2345/v1"
def test_custom_endpoint_uses_config_saved_base_url(self, monkeypatch):
config = {
"model": {
@@ -879,73 +849,9 @@ class TestAuxiliaryPoolAwareness:
class TestGetAuxiliaryProvider:
"""Tests for _get_auxiliary_provider env var resolution."""
def test_no_task_returns_auto(self):
assert _get_auxiliary_provider() == "auto"
assert _get_auxiliary_provider("") == "auto"
def test_auxiliary_prefix_takes_priority(self, monkeypatch):
monkeypatch.setenv("AUXILIARY_VISION_PROVIDER", "openrouter")
assert _get_auxiliary_provider("vision") == "openrouter"
def test_context_prefix_fallback(self, monkeypatch):
monkeypatch.setenv("CONTEXT_COMPRESSION_PROVIDER", "nous")
assert _get_auxiliary_provider("compression") == "nous"
def test_auxiliary_prefix_over_context_prefix(self, monkeypatch):
monkeypatch.setenv("AUXILIARY_COMPRESSION_PROVIDER", "openrouter")
monkeypatch.setenv("CONTEXT_COMPRESSION_PROVIDER", "nous")
assert _get_auxiliary_provider("compression") == "openrouter"
def test_auto_value_treated_as_auto(self, monkeypatch):
monkeypatch.setenv("AUXILIARY_VISION_PROVIDER", "auto")
assert _get_auxiliary_provider("vision") == "auto"
def test_whitespace_stripped(self, monkeypatch):
monkeypatch.setenv("AUXILIARY_VISION_PROVIDER", " openrouter ")
assert _get_auxiliary_provider("vision") == "openrouter"
def test_case_insensitive(self, monkeypatch):
monkeypatch.setenv("AUXILIARY_VISION_PROVIDER", "OpenRouter")
assert _get_auxiliary_provider("vision") == "openrouter"
def test_main_provider(self, monkeypatch):
monkeypatch.setenv("AUXILIARY_WEB_EXTRACT_PROVIDER", "main")
assert _get_auxiliary_provider("web_extract") == "main"
class TestTaskSpecificOverrides:
"""Integration tests for per-task provider routing via get_text_auxiliary_client(task=...)."""
def test_text_with_vision_provider_override(self, monkeypatch):
"""AUXILIARY_VISION_PROVIDER should not affect text tasks."""
monkeypatch.setenv("AUXILIARY_VISION_PROVIDER", "nous")
monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
with patch("agent.auxiliary_client.OpenAI"):
client, model = get_text_auxiliary_client() # no task → auto
assert model == "google/gemini-3-flash-preview" # OpenRouter, not Nous
def test_compression_task_reads_context_prefix(self, monkeypatch):
"""Compression task should check CONTEXT_COMPRESSION_PROVIDER env var."""
monkeypatch.setenv("CONTEXT_COMPRESSION_PROVIDER", "nous")
monkeypatch.setenv("OPENROUTER_API_KEY", "or-key") # would win in auto
with patch("agent.auxiliary_client._read_nous_auth") as mock_nous, \
patch("agent.auxiliary_client.OpenAI"):
mock_nous.return_value = {"access_token": "***"}
client, model = get_text_auxiliary_client("compression")
# Config-first: model comes from config.yaml summary_model default,
# but provider is forced to Nous via env var
assert client is not None
def test_web_extract_task_override(self, monkeypatch):
monkeypatch.setenv("AUXILIARY_WEB_EXTRACT_PROVIDER", "openrouter")
monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
with patch("agent.auxiliary_client.OpenAI"):
client, model = get_text_auxiliary_client("web_extract")
assert model == "google/gemini-3-flash-preview"
def test_task_direct_endpoint_from_config(self, monkeypatch, tmp_path):
hermes_home = tmp_path / "hermes"
hermes_home.mkdir(parents=True, exist_ok=True)
@@ -979,8 +885,6 @@ class TestTaskSpecificOverrides:
"""model:
default: glm-5.1
provider: opencode-go
compression:
summary_provider: auto
"""
)
monkeypatch.setenv("HERMES_HOME", str(hermes_home))
@@ -1039,25 +943,6 @@ model:
"model": "gpt-5.4",
}
def test_compression_summary_base_url_from_config(self, monkeypatch, tmp_path):
"""compression.summary_base_url should produce a custom-endpoint client."""
hermes_home = tmp_path / "hermes"
hermes_home.mkdir(parents=True, exist_ok=True)
(hermes_home / "config.yaml").write_text(
"""compression:
summary_provider: custom
summary_model: glm-4.7
summary_base_url: https://api.z.ai/api/coding/paas/v4
"""
)
monkeypatch.setenv("HERMES_HOME", str(hermes_home))
# Custom endpoints need an API key to build the client
monkeypatch.setenv("OPENAI_API_KEY", "test-key")
with patch("agent.auxiliary_client.OpenAI") as mock_openai:
client, model = get_text_auxiliary_client("compression")
assert model == "glm-4.7"
assert mock_openai.call_args.kwargs["base_url"] == "https://api.z.ai/api/coding/paas/v4"
class TestAuxiliaryMaxTokensParam:
def test_codex_fallback_uses_max_tokens(self, monkeypatch):

View File

@@ -273,18 +273,6 @@ class TestDefaultConfigShape:
assert web["provider"] == "auto"
assert web["model"] == ""
def test_compression_provider_default(self):
from hermes_cli.config import DEFAULT_CONFIG
compression = DEFAULT_CONFIG["compression"]
assert "summary_provider" in compression
assert compression["summary_provider"] == "auto"
def test_compression_base_url_default(self):
from hermes_cli.config import DEFAULT_CONFIG
compression = DEFAULT_CONFIG["compression"]
assert "summary_base_url" in compression
assert compression["summary_base_url"] is None
# ── CLI defaults parity ─────────────────────────────────────────────────────

View File

@@ -12,17 +12,6 @@ def _isolate(tmp_path, monkeypatch):
hermes_home = tmp_path / ".hermes"
hermes_home.mkdir()
monkeypatch.setenv("HERMES_HOME", str(hermes_home))
for env_var in (
"AUXILIARY_VISION_PROVIDER",
"AUXILIARY_VISION_MODEL",
"AUXILIARY_VISION_BASE_URL",
"AUXILIARY_VISION_API_KEY",
"CONTEXT_VISION_PROVIDER",
"CONTEXT_VISION_MODEL",
"CONTEXT_VISION_BASE_URL",
"CONTEXT_VISION_API_KEY",
):
monkeypatch.delenv(env_var, raising=False)
# Write a minimal config so load_config doesn't fail
(hermes_home / "config.yaml").write_text("model:\n default: test-model\n")