fix: ChatContext is not a sequence — use .items for len/list; add test suite

Bug fix:
- GemmaLLM.chat(): len(chat_ctx) → len(chat_ctx.items)
- GemmaLLM._compact_context(): list(chat_ctx) → list(chat_ctx.items)
  ChatContext in livekit-agents 1.7 is not iterable or sized directly;
  it exposes an .items list. The TypeError was silently killing all LLM
  responses (agent heard user but never spoke back).

Test suite (tests/):
- test_web_ui.py: static assets, token endpoint, LiveKit WS proxy
- test_llm_api.py: completion, tool calling, thinking-disabled latency, streaming
- test_mcp_tools.py: weather (Fahrenheit), time, memory, skills — all via
  docker exec + in-container MCP client
- test_agent_integration.py: process alive, worker registered, supervisord,
  container health, LiveKit API
- test_compaction.py: size-triggered compaction, system prompt preservation,
  short-context no-op
- conftest.py: shared fixtures (HTTPS client, token, LLM, docker exec helpers)
- Run with: .venv-tests/bin/python -m pytest tests/ -v

tests/test_mcp_tools.py

@@ -0,0 +1,132 @@
+"""Tests for the MCP tool servers (weather, memory, skills) via docker exec.
+
+Each test spawns the real MCP server script inside the container and drives it
+with a JSON-RPC session using the mcp client library from the agent venv.
+"""
+
+import re
+
+import pytest
+
+
+def _extract_number(text: str) -> float | None:
+    """Pull the first plausible temperature number out of a weather summary."""
+    m = re.search(r"(\d+)\s*degrees", text)
+    if m:
+        return float(m.group(1))
+    m = re.search(r"\b(\d{2,3})\b", text)
+    return float(m.group(1)) if m else None
+
+
+@pytest.mark.slow
+def test_weather_get(mcp_client_factory):
+    text, is_error = mcp_client_factory("weather_mcp.py", "get_weather", {"location": "Raleigh"})
+    assert not is_error
+    assert "degrees" in text.lower()
+
+
+@pytest.mark.slow
+def test_weather_fahrenheit(mcp_client_factory):
+    """wttr.in reports Celsius; the tool must convert to Fahrenheit."""
+    text, is_error = mcp_client_factory("weather_mcp.py", "get_weather", {"location": "Raleigh"})
+    assert not is_error
+    temp = _extract_number(text)
+    assert temp is not None, f"no temperature found in: {text}"
+    # Fahrenheit for any inhabited place is > 50 (Celsius would be ~10-30).
+    assert temp > 50, f"temperature {temp} looks like Celsius, expected Fahrenheit"
+
+
+@pytest.mark.slow
+def test_time_get(mcp_client_factory):
+    text, is_error = mcp_client_factory("weather_mcp.py", "get_time", {})
+    assert not is_error
+    assert ("AM" in text or "PM" in text)
+    days = ("Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday")
+    assert any(d in text for d in days), f"no day name in: {text}"
+
+
+@pytest.mark.slow
+def test_memory_save_and_recall(mcp_client_factory):
+    topic = "test-suite-probe"
+    content = "zebra42 is the probe fact for the voice test suite"
+
+    saved, err = mcp_client_factory(
+        "memory_mcp.py", "memory_save", {"topic": topic, "content": content}
+    )
+    assert not err
+    assert "Saved" in saved
+
+    try:
+        recalled, err = mcp_client_factory("memory_mcp.py", "memory_recall", {"query": "zebra42"})
+        assert not err
+        assert "zebra42" in recalled
+        assert topic in recalled  # recall output is tagged with the topic name
+    finally:
+        # Clean up: remove the probe file so repeated runs stay green.
+        from conftest import docker_exec
+
+        docker_exec("rm", "-f", "/memory/test-suite-probe.md")
+
+
+def test_memory_list(mcp_client_factory):
+    from conftest import docker_exec
+
+    topic = "test-suite-list-probe"
+    try:
+        saved, err = mcp_client_factory(
+            "memory_mcp.py", "memory_save", {"topic": topic, "content": "listing probe"}
+        )
+        assert not err
+        listed, err = mcp_client_factory("memory_mcp.py", "memory_list", {})
+        assert not err
+        assert topic in listed
+    finally:
+        docker_exec("rm", "-f", "/memory/test-suite-list-probe.md")
+
+
+@pytest.mark.slow
+def test_skill_save_and_recall(mcp_client_factory):
+    name = "Test Suite Probe Skill"
+    description = "A probe skill used by the voice test suite."
+    steps = "Step 1: do the thing. Step 2: verify zebra42 was done."
+
+    saved, err = mcp_client_factory(
+        "skills_mcp.py",
+        "skill_save",
+        {"name": name, "description": description, "steps": steps},
+    )
+    assert not err
+    assert "Saved" in saved
+
+    try:
+        recalled, err = mcp_client_factory("skills_mcp.py", "skill_recall", {"query": "zebra42"})
+        assert not err
+        assert "zebra42" in recalled
+        assert "Steps" in recalled
+    finally:
+        from conftest import docker_exec
+
+        docker_exec("rm", "-f", "/skills/test-suite-probe-skill.md")
+
+
+def test_skill_list(mcp_client_factory):
+    from conftest import docker_exec
+
+    name = "Test Suite List Skill"
+    try:
+        saved, err = mcp_client_factory(
+            "skills_mcp.py",
+            "skill_save",
+            {
+                "name": name,
+                "description": "listing probe skill",
+                "steps": "one step",
+            },
+        )
+        assert not err
+        listed, err = mcp_client_factory("skills_mcp.py", "skill_list", {})
+        assert not err
+        # list output uses the slugified filename
+        assert "test-suite-list-skill" in listed
+    finally:
+        docker_exec("rm", "-f", "/skills/test-suite-list-skill.md")