Files
hermes-agent/tests/test_cli_model_command.py

133 lines
5.5 KiB
Python
Raw Normal View History

"""Regression tests for the `/model` slash command in the interactive CLI."""
from unittest.mock import patch, MagicMock
from cli import HermesCLI
class TestModelCommand:
def _make_cli(self):
cli_obj = HermesCLI.__new__(HermesCLI)
cli_obj.model = "anthropic/claude-opus-4.6"
cli_obj.agent = object()
cli_obj.provider = "openrouter"
cli_obj.requested_provider = "openrouter"
cli_obj.base_url = "https://openrouter.ai/api/v1"
cli_obj.api_key = "test-key"
cli_obj._explicit_api_key = None
cli_obj._explicit_base_url = None
return cli_obj
def test_valid_model_from_api_saved_to_config(self, capsys):
cli_obj = self._make_cli()
with patch("hermes_cli.models.fetch_api_models",
return_value=["anthropic/claude-sonnet-4.5", "openai/gpt-5.4"]), \
patch("cli.save_config_value", return_value=True) as save_mock:
cli_obj.process_command("/model anthropic/claude-sonnet-4.5")
output = capsys.readouterr().out
assert "saved to config" in output
assert cli_obj.model == "anthropic/claude-sonnet-4.5"
save_mock.assert_called_once_with("model.default", "anthropic/claude-sonnet-4.5")
fix: stop rejecting unlisted models, accept with warning instead * fix: use session_key instead of chat_id for adapter interrupt lookups monitor_for_interrupt() in _run_agent was using source.chat_id to query the adapter's has_pending_interrupt() and get_pending_message() methods. But the adapter stores interrupt events under build_session_key(source), which produces a different string (e.g. 'agent:main:telegram:dm' vs '123456'). This key mismatch meant the interrupt was never detected through the adapter path, which is the only active interrupt path for all adapter-based platforms (Telegram, Discord, Slack, etc.). The gateway-level interrupt path (in dispatch_message) is unreachable because the adapter intercepts the 2nd message in handle_message() before it reaches dispatch_message(). Result: sending a new message while subagents were running had no effect — the interrupt was silently lost. Fix: replace all source.chat_id references in the interrupt-related code within _run_agent() with the session_key parameter, which matches the adapter's storage keys. Also adds regression tests verifying session_key vs chat_id consistency. * debug: add file-based logging to CLI interrupt path Temporary instrumentation to diagnose why message-based interrupts don't seem to work during subagent execution. Logs to ~/.hermes/interrupt_debug.log (immune to redirect_stdout). Two log points: 1. When Enter handler puts message into _interrupt_queue 2. When chat() reads it and calls agent.interrupt() This will reveal whether the message reaches the queue and whether the interrupt is actually fired. * fix: accept unlisted models with warning instead of rejecting validate_requested_model() previously hard-rejected any model not found in the provider's API listing. This was too aggressive — users on higher plan tiers (e.g. Z.AI Pro/Max) may have access to models not shown in the public listing (like glm-5 on coding endpoints). Changes: - validate_requested_model: accept unlisted models with a warning note instead of blocking. The model is saved to config and used immediately. - Z.AI setup: always offer glm-5 in the model list regardless of whether a coding endpoint was detected. Pro/Max plans support it. - Z.AI setup detection message: softened from 'GLM-5 is not available' to 'GLM-5 may still be available depending on your plan tier'
2026-03-12 16:02:35 -07:00
def test_unlisted_model_accepted_with_warning(self, capsys):
cli_obj = self._make_cli()
with patch("hermes_cli.models.fetch_api_models",
return_value=["anthropic/claude-opus-4.6"]), \
patch("cli.save_config_value") as save_mock:
cli_obj.process_command("/model anthropic/fake-model")
output = capsys.readouterr().out
fix: stop rejecting unlisted models, accept with warning instead * fix: use session_key instead of chat_id for adapter interrupt lookups monitor_for_interrupt() in _run_agent was using source.chat_id to query the adapter's has_pending_interrupt() and get_pending_message() methods. But the adapter stores interrupt events under build_session_key(source), which produces a different string (e.g. 'agent:main:telegram:dm' vs '123456'). This key mismatch meant the interrupt was never detected through the adapter path, which is the only active interrupt path for all adapter-based platforms (Telegram, Discord, Slack, etc.). The gateway-level interrupt path (in dispatch_message) is unreachable because the adapter intercepts the 2nd message in handle_message() before it reaches dispatch_message(). Result: sending a new message while subagents were running had no effect — the interrupt was silently lost. Fix: replace all source.chat_id references in the interrupt-related code within _run_agent() with the session_key parameter, which matches the adapter's storage keys. Also adds regression tests verifying session_key vs chat_id consistency. * debug: add file-based logging to CLI interrupt path Temporary instrumentation to diagnose why message-based interrupts don't seem to work during subagent execution. Logs to ~/.hermes/interrupt_debug.log (immune to redirect_stdout). Two log points: 1. When Enter handler puts message into _interrupt_queue 2. When chat() reads it and calls agent.interrupt() This will reveal whether the message reaches the queue and whether the interrupt is actually fired. * fix: accept unlisted models with warning instead of rejecting validate_requested_model() previously hard-rejected any model not found in the provider's API listing. This was too aggressive — users on higher plan tiers (e.g. Z.AI Pro/Max) may have access to models not shown in the public listing (like glm-5 on coding endpoints). Changes: - validate_requested_model: accept unlisted models with a warning note instead of blocking. The model is saved to config and used immediately. - Z.AI setup: always offer glm-5 in the model list regardless of whether a coding endpoint was detected. Pro/Max plans support it. - Z.AI setup detection message: softened from 'GLM-5 is not available' to 'GLM-5 may still be available depending on your plan tier'
2026-03-12 16:02:35 -07:00
assert "not found" in output or "Model changed" in output
assert cli_obj.model == "anthropic/fake-model" # accepted
fix: stop rejecting unlisted models, accept with warning instead * fix: use session_key instead of chat_id for adapter interrupt lookups monitor_for_interrupt() in _run_agent was using source.chat_id to query the adapter's has_pending_interrupt() and get_pending_message() methods. But the adapter stores interrupt events under build_session_key(source), which produces a different string (e.g. 'agent:main:telegram:dm' vs '123456'). This key mismatch meant the interrupt was never detected through the adapter path, which is the only active interrupt path for all adapter-based platforms (Telegram, Discord, Slack, etc.). The gateway-level interrupt path (in dispatch_message) is unreachable because the adapter intercepts the 2nd message in handle_message() before it reaches dispatch_message(). Result: sending a new message while subagents were running had no effect — the interrupt was silently lost. Fix: replace all source.chat_id references in the interrupt-related code within _run_agent() with the session_key parameter, which matches the adapter's storage keys. Also adds regression tests verifying session_key vs chat_id consistency. * debug: add file-based logging to CLI interrupt path Temporary instrumentation to diagnose why message-based interrupts don't seem to work during subagent execution. Logs to ~/.hermes/interrupt_debug.log (immune to redirect_stdout). Two log points: 1. When Enter handler puts message into _interrupt_queue 2. When chat() reads it and calls agent.interrupt() This will reveal whether the message reaches the queue and whether the interrupt is actually fired. * fix: accept unlisted models with warning instead of rejecting validate_requested_model() previously hard-rejected any model not found in the provider's API listing. This was too aggressive — users on higher plan tiers (e.g. Z.AI Pro/Max) may have access to models not shown in the public listing (like glm-5 on coding endpoints). Changes: - validate_requested_model: accept unlisted models with a warning note instead of blocking. The model is saved to config and used immediately. - Z.AI setup: always offer glm-5 in the model list regardless of whether a coding endpoint was detected. Pro/Max plans support it. - Z.AI setup detection message: softened from 'GLM-5 is not available' to 'GLM-5 may still be available depending on your plan tier'
2026-03-12 16:02:35 -07:00
def test_api_unreachable_accepts_and_persists(self, capsys):
cli_obj = self._make_cli()
with patch("hermes_cli.models.fetch_api_models", return_value=None), \
patch("cli.save_config_value") as save_mock:
cli_obj.process_command("/model anthropic/claude-sonnet-next")
output = capsys.readouterr().out
fix: stop rejecting unlisted models, accept with warning instead * fix: use session_key instead of chat_id for adapter interrupt lookups monitor_for_interrupt() in _run_agent was using source.chat_id to query the adapter's has_pending_interrupt() and get_pending_message() methods. But the adapter stores interrupt events under build_session_key(source), which produces a different string (e.g. 'agent:main:telegram:dm' vs '123456'). This key mismatch meant the interrupt was never detected through the adapter path, which is the only active interrupt path for all adapter-based platforms (Telegram, Discord, Slack, etc.). The gateway-level interrupt path (in dispatch_message) is unreachable because the adapter intercepts the 2nd message in handle_message() before it reaches dispatch_message(). Result: sending a new message while subagents were running had no effect — the interrupt was silently lost. Fix: replace all source.chat_id references in the interrupt-related code within _run_agent() with the session_key parameter, which matches the adapter's storage keys. Also adds regression tests verifying session_key vs chat_id consistency. * debug: add file-based logging to CLI interrupt path Temporary instrumentation to diagnose why message-based interrupts don't seem to work during subagent execution. Logs to ~/.hermes/interrupt_debug.log (immune to redirect_stdout). Two log points: 1. When Enter handler puts message into _interrupt_queue 2. When chat() reads it and calls agent.interrupt() This will reveal whether the message reaches the queue and whether the interrupt is actually fired. * fix: accept unlisted models with warning instead of rejecting validate_requested_model() previously hard-rejected any model not found in the provider's API listing. This was too aggressive — users on higher plan tiers (e.g. Z.AI Pro/Max) may have access to models not shown in the public listing (like glm-5 on coding endpoints). Changes: - validate_requested_model: accept unlisted models with a warning note instead of blocking. The model is saved to config and used immediately. - Z.AI setup: always offer glm-5 in the model list regardless of whether a coding endpoint was detected. Pro/Max plans support it. - Z.AI setup detection message: softened from 'GLM-5 is not available' to 'GLM-5 may still be available depending on your plan tier'
2026-03-12 16:02:35 -07:00
assert "saved to config" in output
assert cli_obj.model == "anthropic/claude-sonnet-next"
fix: stop rejecting unlisted models, accept with warning instead * fix: use session_key instead of chat_id for adapter interrupt lookups monitor_for_interrupt() in _run_agent was using source.chat_id to query the adapter's has_pending_interrupt() and get_pending_message() methods. But the adapter stores interrupt events under build_session_key(source), which produces a different string (e.g. 'agent:main:telegram:dm' vs '123456'). This key mismatch meant the interrupt was never detected through the adapter path, which is the only active interrupt path for all adapter-based platforms (Telegram, Discord, Slack, etc.). The gateway-level interrupt path (in dispatch_message) is unreachable because the adapter intercepts the 2nd message in handle_message() before it reaches dispatch_message(). Result: sending a new message while subagents were running had no effect — the interrupt was silently lost. Fix: replace all source.chat_id references in the interrupt-related code within _run_agent() with the session_key parameter, which matches the adapter's storage keys. Also adds regression tests verifying session_key vs chat_id consistency. * debug: add file-based logging to CLI interrupt path Temporary instrumentation to diagnose why message-based interrupts don't seem to work during subagent execution. Logs to ~/.hermes/interrupt_debug.log (immune to redirect_stdout). Two log points: 1. When Enter handler puts message into _interrupt_queue 2. When chat() reads it and calls agent.interrupt() This will reveal whether the message reaches the queue and whether the interrupt is actually fired. * fix: accept unlisted models with warning instead of rejecting validate_requested_model() previously hard-rejected any model not found in the provider's API listing. This was too aggressive — users on higher plan tiers (e.g. Z.AI Pro/Max) may have access to models not shown in the public listing (like glm-5 on coding endpoints). Changes: - validate_requested_model: accept unlisted models with a warning note instead of blocking. The model is saved to config and used immediately. - Z.AI setup: always offer glm-5 in the model list regardless of whether a coding endpoint was detected. Pro/Max plans support it. - Z.AI setup detection message: softened from 'GLM-5 is not available' to 'GLM-5 may still be available depending on your plan tier'
2026-03-12 16:02:35 -07:00
save_mock.assert_called_once()
fix: stop rejecting unlisted models, accept with warning instead * fix: use session_key instead of chat_id for adapter interrupt lookups monitor_for_interrupt() in _run_agent was using source.chat_id to query the adapter's has_pending_interrupt() and get_pending_message() methods. But the adapter stores interrupt events under build_session_key(source), which produces a different string (e.g. 'agent:main:telegram:dm' vs '123456'). This key mismatch meant the interrupt was never detected through the adapter path, which is the only active interrupt path for all adapter-based platforms (Telegram, Discord, Slack, etc.). The gateway-level interrupt path (in dispatch_message) is unreachable because the adapter intercepts the 2nd message in handle_message() before it reaches dispatch_message(). Result: sending a new message while subagents were running had no effect — the interrupt was silently lost. Fix: replace all source.chat_id references in the interrupt-related code within _run_agent() with the session_key parameter, which matches the adapter's storage keys. Also adds regression tests verifying session_key vs chat_id consistency. * debug: add file-based logging to CLI interrupt path Temporary instrumentation to diagnose why message-based interrupts don't seem to work during subagent execution. Logs to ~/.hermes/interrupt_debug.log (immune to redirect_stdout). Two log points: 1. When Enter handler puts message into _interrupt_queue 2. When chat() reads it and calls agent.interrupt() This will reveal whether the message reaches the queue and whether the interrupt is actually fired. * fix: accept unlisted models with warning instead of rejecting validate_requested_model() previously hard-rejected any model not found in the provider's API listing. This was too aggressive — users on higher plan tiers (e.g. Z.AI Pro/Max) may have access to models not shown in the public listing (like glm-5 on coding endpoints). Changes: - validate_requested_model: accept unlisted models with a warning note instead of blocking. The model is saved to config and used immediately. - Z.AI setup: always offer glm-5 in the model list regardless of whether a coding endpoint was detected. Pro/Max plans support it. - Z.AI setup detection message: softened from 'GLM-5 is not available' to 'GLM-5 may still be available depending on your plan tier'
2026-03-12 16:02:35 -07:00
def test_no_slash_model_accepted_with_warning(self, capsys):
cli_obj = self._make_cli()
with patch("hermes_cli.models.fetch_api_models",
return_value=["openai/gpt-5.4"]) as fetch_mock, \
patch("cli.save_config_value") as save_mock:
cli_obj.process_command("/model gpt-5.4")
output = capsys.readouterr().out
# Auto-detection remaps bare model names to proper OpenRouter slugs
assert cli_obj.model == "openai/gpt-5.4"
def test_validation_crash_falls_back_to_save(self, capsys):
cli_obj = self._make_cli()
with patch("hermes_cli.models.validate_requested_model",
side_effect=RuntimeError("boom")), \
patch("cli.save_config_value", return_value=True) as save_mock:
cli_obj.process_command("/model anthropic/claude-sonnet-4.5")
output = capsys.readouterr().out
assert "saved to config" in output
assert cli_obj.model == "anthropic/claude-sonnet-4.5"
save_mock.assert_called_once()
def test_show_model_when_no_argument(self, capsys):
cli_obj = self._make_cli()
cli_obj.process_command("/model")
output = capsys.readouterr().out
assert "anthropic/claude-opus-4.6" in output
assert "OpenRouter" in output
assert "Authenticated providers" in output or "Switch model" in output
assert "provider" in output and "model" in output
# -- provider switching tests -------------------------------------------
def test_provider_colon_model_switches_provider(self, capsys):
cli_obj = self._make_cli()
with patch("hermes_cli.runtime_provider.resolve_runtime_provider", return_value={
"provider": "zai",
"api_key": "zai-key",
"base_url": "https://api.z.ai/api/paas/v4",
}), \
patch("hermes_cli.models.fetch_api_models",
return_value=["glm-5", "glm-4.7"]), \
patch("cli.save_config_value", return_value=True) as save_mock:
cli_obj.process_command("/model zai:glm-5")
output = capsys.readouterr().out
assert "glm-5" in output
assert "provider:" in output.lower() or "Z.AI" in output
assert cli_obj.model == "glm-5"
assert cli_obj.provider == "zai"
assert cli_obj.base_url == "https://api.z.ai/api/paas/v4"
feat(model): /model command overhaul — Phases 2, 3, 5 * feat(model): persist base_url on /model switch, auto-detect for bare /model custom Phase 2+3 of the /model command overhaul: Phase 2 — Persist base_url on model switch: - CLI: save model.base_url when switching to a non-OpenRouter endpoint; clear it when switching away from custom to prevent stale URLs leaking into the new provider's resolution - Gateway: same logic using direct YAML write Phase 3 — Better feedback and edge cases: - Bare '/model custom' now auto-detects the model from the endpoint using _auto_detect_local_model() and saves all three config values (model, provider, base_url) atomically - Shows endpoint URL in success messages when switching to/from custom providers (both CLI and gateway) - Clear error messages when no custom endpoint is configured - Updated test assertions for the additional save_config_value call Fixes #2562 (Phase 2+3) * feat(model): support custom:name:model triple syntax for named custom providers Phase 5 of the /model command overhaul. Extends parse_model_input() to handle the triple syntax: /model custom:local-server:qwen → provider='custom:local-server', model='qwen' /model custom:my-model → provider='custom', model='my-model' (unchanged) The 'custom:local-server' provider string is already supported by _get_named_custom_provider() in runtime_provider.py, which matches it against the custom_providers list in config.yaml. This just wires the parsing so users can do it from the /model slash command. Added 4 tests covering single, triple, whitespace, and empty model cases.
2026-03-24 06:58:04 -07:00
# Model, provider, and base_url should be saved
assert save_mock.call_count == 3
save_calls = [c.args for c in save_mock.call_args_list]
assert ("model.default", "glm-5") in save_calls
assert ("model.provider", "zai") in save_calls
# base_url is also persisted on provider change (Phase 2 fix)
assert any(c[0] == "model.base_url" for c in save_calls)
def test_provider_switch_fails_on_bad_credentials(self, capsys):
cli_obj = self._make_cli()
with patch("hermes_cli.runtime_provider.resolve_runtime_provider",
side_effect=Exception("No API key found")):
cli_obj.process_command("/model nous:hermes-3")
output = capsys.readouterr().out
assert "Could not resolve credentials" in output
assert cli_obj.model == "anthropic/claude-opus-4.6" # unchanged
assert cli_obj.provider == "openrouter" # unchanged