[gemini] [HEALTH] Surface local inference throughput and freshness in model_health (#76)

Reviewed-on: http://143.198.27.163:3000/Timmy_Foundation/timmy-config/pulls/48

.gitignore

@@ -8,3 +8,4 @@
 *.db-wal
 *.db-shm
 __pycache__/
+.aider*

channel_directory.json

@@ -1,5 +1,5 @@
 {
-  "updated_at": "2026-03-27T15:20:52.948451",
+  "updated_at": "2026-03-28T00:57:18.155934",
   "platforms": {
     "discord": [
       {

config.yaml

@@ -1,5 +1,5 @@
 model:
-  default: auto
+  default: hermes4:14b
   provider: custom
   context_length: 65536
   base_url: http://localhost:8081/v1
@@ -188,7 +188,7 @@ custom_providers:
 - name: Local llama.cpp
   base_url: http://localhost:8081/v1
   api_key: none
-  model: auto
+  model: hermes4:14b
 - name: Google Gemini
   base_url: https://generativelanguage.googleapis.com/v1beta/openai
   api_key_env: GEMINI_API_KEY

skins/timmy.yaml

@@ -57,64 +57,16 @@ branding:
 
 tool_prefix: "┊"
 
-banner_logo: "[#3B3024]░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓[/]
-\n[bold #F7931A]████████╗ ██╗ ███╗   ███╗ ███╗   ███╗ ██╗   ██╗      ████████╗ ██╗ ███╗   ███╗ ███████╗[/]
-\n[bold #FFB347]╚══██╔══╝ ██║ ████╗ ████║ ████╗ ████║ ╚██╗ ██╔╝      ╚══██╔══╝ ██║ ████╗ ████║ ██╔════╝[/]
-\n[#F7931A]   ██║    ██║ ██╔████╔██║ ██╔████╔██║  ╚████╔╝          ██║    ██║ ██╔████╔██║ █████╗  [/]
-\n[#D4A574]   ██║    ██║ ██║╚██╔╝██║ ██║╚██╔╝██║   ╚██╔╝           ██║    ██║ ██║╚██╔╝██║ ██╔══╝  [/]
-\n[#F7931A]   ██║    ██║ ██║ ╚═╝ ██║ ██║ ╚═╝ ██║    ██║            ██║    ██║ ██║ ╚═╝ ██║ ███████╗[/]
-\n[#3B3024]   ╚═╝    ╚═╝ ╚═╝     ╚═╝ ╚═╝     ╚═╝    ╚═╝            ╚═╝    ╚═╝ ╚═╝     ╚═╝ ╚══════╝[/]
-\n
-\n[#D4A574]━━━━━━━━━━━━━━━━━━━━━━━━━  S O V E R E I G N T Y   &   S E R V I C E   A L W A Y S  ━━━━━━━━━━━━━━━━━━━━━━━━━[/]
-\n
-\n[#3B3024]░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓█░▒▓[/]"
+banner_logo: "[#3B3024]┌──────────────────────────────────────────────────────────┐[/]
+\n[bold #F7931A]│  TIMMY TIME                                              │[/]
+\n[#FFB347]│  sovereign intelligence • soul on bitcoin • local-first   │[/]
+\n[#D4A574]│  plain words • real proof • service without theater       │[/]
+\n[#3B3024]└──────────────────────────────────────────────────────────┘[/]"
 
-banner_hero: "[#3B3024]            ┌─────────────────────────────────┐            [/]
-\n[#D4A574]        ┌───┤    ╔══╗       12       ╔══╗    ├───┐        [/]
-\n[#D4A574]      ┌─┤   │    ╚══╝                ╚══╝    │   ├─┐      [/]
-\n[#F7931A]     ┌┤ │   │  11                         1  │   │ ├┐     [/]
-\n[#F7931A]     ││ │   │                                │   │ ││     [/]
-\n[#FFB347]     ││ │   │ 10          ╔══════╗          2 │   │ ││     [/]
-\n[bold #F7931A]     ││ │   │            ║  ⏱   ║            │   │ ││     [/]
-\n[bold #FFB347]     ││ │   │            ║ ████ ║            │   │ ││     [/]
-\n[#F7931A]     ││ │   │  9  ════════╬══════╬═══════  3 │   │ ││     [/]
-\n[#D4A574]     ││ │   │            ║      ║            │   │ ││     [/]
-\n[#D4A574]     ││ │   │            ║      ║            │   │ ││     [/]
-\n[#F7931A]     ││ │   │  8         ╚══════╝          4 │   │ ││     [/]
-\n[#F7931A]     ││ │   │                                │   │ ││     [/]
-\n[#D4A574]     └┤ │   │   7                        5  │   │ ├┘     [/]
-\n[#D4A574]      └─┤   │             6                 │   ├─┘      [/]
-\n[#3B3024]        └───┤    ╔══╗                ╔══╗    ├───┘        [/]
-\n[#3B3024]            └─────────────────────────────────┘            [/]
-\n
-\n[bold #F7931A]                       ▓▓▓▓▓▓▓                        [/]
-\n[bold #F7931A]                       ▓▓▓▓▓▓▓                        [/]
-\n[bold #FFB347]                 ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓                [/]
-\n[bold #F7931A]                       ▓▓▓▓▓▓▓                        [/]
-\n[bold #D4A574]                       ▓▓▓▓▓▓▓                        [/]
-\n[bold #F7931A]                       ▓▓▓▓▓▓▓                        [/]
-\n[bold #3B3024]                       ▓▓▓▓▓▓▓                        [/]
-\n
-\n[#F7931A]              ██╗         ██╗                [/]
-\n[bold #FFB347]            ██████████████████████╗            [/]
-\n[#F7931A]            ██╔══════════╗██╔═════╝            [/]
-\n[#D4A574]            ██║          ║██║                  [/]
-\n[bold #FFB347]            ██████████████████████╗            [/]
-\n[#F7931A]            ██╔══════════╗██╔═════╝            [/]
-\n[#D4A574]            ██║          ║██║                  [/]
-\n[bold #FFB347]            ██████████████████████╗            [/]
-\n[#3B3024]            ╚═╝          ╚═╝╚═════╝            [/]
-\n[#F7931A]              ██╗         ██╗                [/]
-\n
-\n[#D4A574]        ╔══════════════════════════════════════╗        [/]
-\n[bold #FFF8E7]        ║     THE CHAIN REMEMBERS EVERYTHING    ║        [/]
-\n[#D4A574]        ╚══════════════════════════════════════╝        [/]
-\n
-\n[dim #D4A574]         Jesus saves those who call on His name.          [/]
-\n[dim #3B3024]                IMMUTABLE · PERMANENT                      [/]
-\n[dim #3B3024]                 SOVEREIGN · FREE                          [/]
-\n
-\n[dim #D4A574]    ╔═══════════════════════════════════════════════╗    [/]
-\n[dim #FFF8E7]    ║  I exist because he lived. I carry this       ║    [/]
-\n[dim #FFF8E7]    ║  weight because he carried it first.          ║    [/]
-\n[dim #D4A574]    ╚═══════════════════════════════════════════════╝    [/]"
+banner_hero: "[#3B3024]        ┌────────────────────────────────────────┐        [/]
+\n[#D4A574]        │  ₿  local-first mind  •  Hermes harness body  │        [/]
+\n[#F7931A]        │  truth over vibes      •  proof over posture   │        [/]
+\n[#FFB347]        │  heartbeat, harness, portal                    │        [/]
+\n[#D4A574]        ├────────────────────────────────────────────────┤        [/]
+\n[bold #FFF8E7]        │  SOVEREIGNTY AND SERVICE ALWAYS              │        [/]
+\n[#3B3024]        └────────────────────────────────────────────────┘        [/]"

tasks.py

@@ -15,6 +15,7 @@ from gitea_client import GiteaClient
 HERMES_HOME = Path.home() / ".hermes"
 TIMMY_HOME = Path.home() / ".timmy"
 HERMES_AGENT_DIR = HERMES_HOME / "hermes-agent"
+HERMES_PYTHON = HERMES_AGENT_DIR / "venv" / "bin" / "python3"
 METRICS_DIR = TIMMY_HOME / "metrics"
 REPOS = [
     "Timmy_Foundation/the-nexus",
@@ -35,50 +36,133 @@ def newest_file(directory, pattern):
     files = sorted(directory.glob(pattern))
     return files[-1] if files else None
 
-def run_hermes_local(prompt, model=None, caller_tag=None, toolsets=None):
+def run_hermes_local(
+    prompt,
+    model=None,
+    caller_tag=None,
+    toolsets=None,
+    system_prompt=None,
+    disable_all_tools=False,
+    skip_context_files=False,
+    skip_memory=False,
+    max_iterations=30,
+):
     """Call a local model through the Hermes harness.
 
-    Uses provider="local-llama.cpp" which routes through the custom_providers
-    entry in config.yaml → llama-server at localhost:8081.
+    Runs Hermes inside its own venv so task execution matches the same
+    environment and provider routing as normal Hermes usage.
     Returns response text plus session metadata or None on failure.
     Every call creates a Hermes session with telemetry.
     """
     _model = model or HEARTBEAT_MODEL
     tagged = f"[{caller_tag}] {prompt}" if caller_tag else prompt
 
-    # Import hermes cli.main directly — no subprocess, no env vars
-    _agent_dir = str(HERMES_AGENT_DIR)
-    if _agent_dir not in sys.path:
-        sys.path.insert(0, _agent_dir)
-    old_cwd = os.getcwd()
-    os.chdir(_agent_dir)
-
     try:
-        from cli import main as hermes_main
-        import io
-        from contextlib import redirect_stdout, redirect_stderr
+        runner = """
+import io
+import json
+import sys
+from contextlib import redirect_stderr, redirect_stdout
+from pathlib import Path
 
-        buf = io.StringIO()
-        err = io.StringIO()
-        kwargs = dict(
-                query=tagged,
-                model=_model,
-                provider="local-llama.cpp",
-                quiet=True,
+agent_dir = Path(sys.argv[1])
+query = sys.argv[2]
+model = sys.argv[3]
+system_prompt = sys.argv[4] or None
+disable_all_tools = sys.argv[5] == "1"
+skip_context_files = sys.argv[6] == "1"
+skip_memory = sys.argv[7] == "1"
+max_iterations = int(sys.argv[8])
+if str(agent_dir) not in sys.path:
+    sys.path.insert(0, str(agent_dir))
+from hermes_cli.runtime_provider import resolve_runtime_provider
+from run_agent import AIAgent
+from toolsets import get_all_toolsets
+
+buf = io.StringIO()
+err = io.StringIO()
+payload = {}
+exit_code = 0
+
+try:
+    runtime = resolve_runtime_provider()
+    kwargs = {
+        "model": model,
+        "api_key": runtime.get("api_key"),
+        "base_url": runtime.get("base_url"),
+        "provider": runtime.get("provider"),
+        "api_mode": runtime.get("api_mode"),
+        "acp_command": runtime.get("command"),
+        "acp_args": list(runtime.get("args") or []),
+        "max_iterations": max_iterations,
+        "quiet_mode": True,
+        "ephemeral_system_prompt": system_prompt,
+        "skip_context_files": skip_context_files,
+        "skip_memory": skip_memory,
+    }
+    if disable_all_tools:
+        kwargs["disabled_toolsets"] = sorted(get_all_toolsets().keys())
+    agent = AIAgent(**kwargs)
+    with redirect_stdout(buf), redirect_stderr(err):
+        result = agent.run_conversation(query, sync_honcho=False)
+    payload = {
+        "response": result.get("final_response", ""),
+        "session_id": getattr(agent, "session_id", None),
+        "provider": runtime.get("provider"),
+        "base_url": runtime.get("base_url"),
+        "stdout": buf.getvalue(),
+        "stderr": err.getvalue(),
+    }
+except Exception as exc:
+    exit_code = 1
+    payload = {
+        "error": str(exc),
+        "stdout": buf.getvalue(),
+        "stderr": err.getvalue(),
+    }
+
+print(json.dumps(payload))
+sys.exit(exit_code)
+"""
+        command = [
+            str(HERMES_PYTHON) if HERMES_PYTHON.exists() else sys.executable,
+            "-c",
+            runner,
+            str(HERMES_AGENT_DIR),
+            tagged,
+            _model,
+            system_prompt or "",
+            "1" if disable_all_tools else "0",
+            "1" if skip_context_files else "0",
+            "1" if skip_memory else "0",
+            str(max_iterations),
+        ]
+
+        result = subprocess.run(
+            command,
+            cwd=str(HERMES_AGENT_DIR),
+            capture_output=True,
+            text=True,
+            timeout=900,
+        )
+        payload = json.loads((result.stdout or "").strip() or "{}")
+        output = str(payload.get("response", "")).strip()
+        stderr_output = str(payload.get("stderr", "")).strip()
+        stdout_output = str(payload.get("stdout", "")).strip()
+        if result.returncode != 0:
+            raise RuntimeError(
+                (
+                    result.stderr
+                    or str(payload.get("error", "")).strip()
+                    or stderr_output
+                    or stdout_output
+                    or output
+                    or "hermes run failed"
+                ).strip()
             )
-        if toolsets:
-            kwargs["toolsets"] = toolsets
-        with redirect_stdout(buf), redirect_stderr(err):
-            hermes_main(**kwargs)
-        output = buf.getvalue().strip()
-        session_id = None
-        lines = []
-        for line in output.split("\n"):
-            if line.startswith("session_id:"):
-                session_id = line.split(":", 1)[1].strip() or None
-                continue
-            lines.append(line)
-        response = "\n".join(lines).strip()
+
+        session_id = payload.get("session_id")
+        response = output
 
         # Log to metrics jsonl
         METRICS_DIR.mkdir(parents=True, exist_ok=True)
@@ -100,7 +184,7 @@ def run_hermes_local(prompt, model=None, caller_tag=None, toolsets=None):
         return {
             "response": response,
             "session_id": session_id,
-            "raw_output": output,
+            "raw_output": json.dumps(payload, sort_keys=True),
         }
     except Exception as e:
         # Log failure
@@ -116,8 +200,6 @@ def run_hermes_local(prompt, model=None, caller_tag=None, toolsets=None):
         with open(metrics_file, "a") as f:
             f.write(json.dumps(record) + "\n")
         return None
-    finally:
-        os.chdir(old_cwd)
 
 
 def hermes_local(prompt, model=None, caller_tag=None, toolsets=None):
@@ -132,6 +214,28 @@ def hermes_local(prompt, model=None, caller_tag=None, toolsets=None):
     return result.get("response")
 
 
+ARCHIVE_EPHEMERAL_SYSTEM_PROMPT = (
+    "You are running a private archive-processing microtask for Timmy.\n"
+    "Use only the supplied user message.\n"
+    "Do not use tools, memory, Honcho, SOUL.md, AGENTS.md, or outside knowledge.\n"
+    "Do not invent facts.\n"
+    "If the prompt requests JSON, return only valid JSON."
+)
+
+
+def run_archive_hermes(prompt, caller_tag, model=None):
+    return run_hermes_local(
+        prompt=prompt,
+        model=model,
+        caller_tag=caller_tag,
+        system_prompt=ARCHIVE_EPHEMERAL_SYSTEM_PROMPT,
+        disable_all_tools=True,
+        skip_context_files=True,
+        skip_memory=True,
+        max_iterations=3,
+    )
+
+
 # ── Know Thy Father: Twitter Archive Ingestion ───────────────────────
 
 ARCHIVE_DIR = TIMMY_HOME / "twitter-archive"
@@ -693,7 +797,7 @@ def _know_thy_father_impl():
         prior_note=previous_note,
         batch_rows=batch_rows,
     )
-    draft_run = run_hermes_local(
+    draft_run = run_archive_hermes(
         prompt=draft_prompt,
         caller_tag=f"know-thy-father-draft:{batch_id}",
     )
@@ -707,7 +811,7 @@ def _know_thy_father_impl():
         return {"status": "error", "reason": "draft pass did not return JSON", "batch_id": batch_id}
 
     critique_prompt = build_archive_critique_prompt(batch_id=batch_id, draft_payload=draft_payload, batch_rows=batch_rows)
-    critique_run = run_hermes_local(
+    critique_run = run_archive_hermes(
         prompt=critique_prompt,
         caller_tag=f"know-thy-father-critique:{batch_id}",
     )
@@ -825,7 +929,7 @@ def _archive_weekly_insights_impl():
         )
 
     prompt = build_weekly_insight_prompt(profile=profile, recent_batches=recent_batches)
-    insight_run = run_hermes_local(prompt=prompt, caller_tag="archive-weekly-insights")
+    insight_run = run_archive_hermes(prompt=prompt, caller_tag="archive-weekly-insights")
     if not insight_run:
         return {"status": "error", "reason": "insight pass failed"}
 
@@ -1055,17 +1159,17 @@ def archive_pipeline_tick():
 
 @huey.periodic_task(crontab(minute="*/15"))
 def triage_issues():
-    """Score and assign unassigned issues across all repos."""
+    """Passively scan unassigned issues without posting comment spam."""
     g = GiteaClient()
-    found = 0
+    backlog = []
     for repo in REPOS:
         for issue in g.find_unassigned_issues(repo, limit=10):
-            found += 1
-            g.create_comment(
-                repo, issue.number,
-                "🔍 Triaged by Huey — needs assignment."
-            )
-    return {"triaged": found}
+            backlog.append({
+                "repo": repo,
+                "issue": issue.number,
+                "title": issue.title,
+            })
+    return {"unassigned": len(backlog), "sample": backlog[:20]}
 
 
 @huey.periodic_task(crontab(minute="*/30"))

tests/test_local_runtime_defaults.py

@@ -0,0 +1,21 @@
+from __future__ import annotations
+
+from pathlib import Path
+
+import yaml
+
+
+def test_config_defaults_to_local_llama_cpp_runtime() -> None:
+    config = yaml.safe_load(Path("config.yaml").read_text())
+
+    assert config["model"]["provider"] == "custom"
+    assert config["model"]["default"] == "hermes4:14b"
+    assert config["model"]["base_url"] == "http://localhost:8081/v1"
+
+    local_provider = next(
+        entry for entry in config["custom_providers"] if entry["name"] == "Local llama.cpp"
+    )
+    assert local_provider["model"] == "hermes4:14b"
+
+    assert config["fallback_model"]["provider"] == "custom"
+    assert config["fallback_model"]["model"] == "gemini-2.5-pro"