fix: harden auxiliary model config — gateway bridge, vision safety, tests

Improvements on top of PR #606 (auxiliary model configuration):

1. Gateway bridge: Added auxiliary.* and compression.summary_provider
   config bridging to gateway/run.py so config.yaml settings work from
   messaging platforms (not just CLI). Matches the pattern in cli.py.

2. Vision auto-fallback safety: In auto mode, vision now only tries
   OpenRouter + Nous Portal (known multimodal-capable providers).
   Custom endpoints, Codex, and API-key providers are skipped to avoid
   confusing errors from providers that don't support vision input.
   Explicit provider override (AUXILIARY_VISION_PROVIDER=main) still
   allows using any provider.

3. Comprehensive tests (46 new):
   - _get_auxiliary_provider env var resolution (8 tests)
   - _resolve_forced_provider with all provider types (8 tests)
   - Per-task provider routing integration (4 tests)
   - Vision auto-fallback safety (7 tests)
   - Config bridging logic (11 tests)
   - Gateway/CLI bridge parity (2 tests)
   - Vision model override via env var (2 tests)
   - DEFAULT_CONFIG shape validation (4 tests)

4. Docs: Added auxiliary_client.py to AGENTS.md project structure.
   Updated module docstring with separate text/vision resolution chains.

Tests: 2429 passed (was 2383).
This commit is contained in:
teknium1 2026-03-08 18:06:40 -07:00
parent d9f373654b
commit 5ae0b731d0
5 changed files with 534 additions and 5 deletions

View file

@ -86,10 +86,29 @@ if _config_path.exists():
"enabled": "CONTEXT_COMPRESSION_ENABLED",
"threshold": "CONTEXT_COMPRESSION_THRESHOLD",
"summary_model": "CONTEXT_COMPRESSION_MODEL",
"summary_provider": "CONTEXT_COMPRESSION_PROVIDER",
}
for _cfg_key, _env_var in _compression_env_map.items():
if _cfg_key in _compression_cfg:
os.environ[_env_var] = str(_compression_cfg[_cfg_key])
# Auxiliary model overrides (vision, web_extract).
# Each task has provider + model; bridge non-default values to env vars.
_auxiliary_cfg = _cfg.get("auxiliary", {})
if _auxiliary_cfg and isinstance(_auxiliary_cfg, dict):
_aux_task_env = {
"vision": ("AUXILIARY_VISION_PROVIDER", "AUXILIARY_VISION_MODEL"),
"web_extract": ("AUXILIARY_WEB_EXTRACT_PROVIDER", "AUXILIARY_WEB_EXTRACT_MODEL"),
}
for _task_key, (_prov_env, _model_env) in _aux_task_env.items():
_task_cfg = _auxiliary_cfg.get(_task_key, {})
if not isinstance(_task_cfg, dict):
continue
_prov = str(_task_cfg.get("provider", "")).strip()
_model = str(_task_cfg.get("model", "")).strip()
if _prov and _prov != "auto":
os.environ[_prov_env] = _prov
if _model:
os.environ[_model_env] = _model
_agent_cfg = _cfg.get("agent", {})
if _agent_cfg and isinstance(_agent_cfg, dict):
if "max_turns" in _agent_cfg: