feat: add fleet cost report for #520

scripts/README_bezalel_gemma4_vps.md

@@ -1,51 +0,0 @@
-# Bezalel Gemma 4 VPS Wiring
-
-Issue: timmy-home #544
-
-This helper is the repo-side operator bundle for wiring a live Gemma 4 endpoint into Bezalel's VPS config without hardcoding one dead pod forever.
-
-What `scripts/bezalel_gemma4_vps.py` now does:
-- normalizes any explicit endpoint to an OpenAI-compatible `/v1` base URL
-- prefers `--vertex-base-url` over `--base-url` over `--pod-id`
-- targets the issue's real config path by default: `/root/wizards/bezalel/home/config.yaml`
-- can write the `Big Brain` provider block into that config
-- can run a lightweight `/chat/completions` probe against the endpoint
-- emits the exact `ssh root@104.131.15.18 ... curl ...` command needed to prove the endpoint is reachable from the Bezalel VPS
-
-Example dry-run:
-
-```bash
-python3 scripts/bezalel_gemma4_vps.py \
-  --base-url https://<pod-id>-11434.proxy.runpod.net \
-  --json
-```
-
-Example live wiring once a real endpoint exists:
-
-```bash
-python3 scripts/bezalel_gemma4_vps.py \
-  --base-url https://<pod-id>-11434.proxy.runpod.net \
-  --config-path /root/wizards/bezalel/home/config.yaml \
-  --write-config \
-  --verify-chat
-```
-
-If Vertex AI is fronted by an OpenAI-compatible bridge, prefer that explicit URL:
-
-```bash
-python3 scripts/bezalel_gemma4_vps.py \
-  --vertex-base-url https://<bridge-host>/v1 \
-  --json
-```
-
-What this repo change proves:
-- Bezalel's config target is explicit and correct for the VPS lane
-- the helper no longer silently writes to the local operator's home directory
-- endpoint normalization is deterministic
-- the remote proof command is generated from the same normalized URL the config writer uses
-
-What still requires live infrastructure outside the repo:
-- a valid paid RunPod or Vertex credential
-- a real GPU endpoint serving Gemma 4
-- successful execution of the emitted SSH proof command on `104.131.15.18`
-- successful Bezalel Hermes chat against that live endpoint

scripts/bezalel_gemma4_vps.py

@@ -8,14 +8,12 @@ Safe by default:
 - can call the RunPod GraphQL API if a key is provided and --apply-runpod is used
 - can update a Hermes config file in-place when --write-config is used
 - can verify an OpenAI-compatible endpoint with a lightweight chat probe
-- emits the exact Bezalel VPS curl proof command for remote verification
 """
 
 from __future__ import annotations
 
 import argparse
 import json
-import shlex
 from pathlib import Path
 from typing import Any
 from urllib import request
@@ -29,9 +27,7 @@ DEFAULT_IMAGE = "ollama/ollama:latest"
 DEFAULT_MODEL = "gemma4:latest"
 DEFAULT_PROVIDER_NAME = "Big Brain"
 DEFAULT_TOKEN_FILE = Path.home() / ".config" / "runpod" / "access_key"
-DEFAULT_CONFIG_PATH = Path("/root/wizards/bezalel/home/config.yaml")
-DEFAULT_BEZALEL_VPS_HOST = "104.131.15.18"
-DEFAULT_VERIFY_PROMPT = "Say READY"
+DEFAULT_CONFIG_PATH = Path.home() / "wizards" / "bezalel" / "home" / "config.yaml"
 
 
 def build_deploy_mutation(
@@ -67,31 +63,8 @@ mutation {{
 '''.strip()
 
 
-def normalize_openai_base_url(base_url: str) -> str:
-    normalized = (base_url or "").strip().rstrip("/")
-    if not normalized:
-        return normalized
-    for suffix in ("/chat/completions", "/models"):
-        if normalized.endswith(suffix):
-            normalized = normalized[: -len(suffix)]
-            break
-    if not normalized.endswith("/v1"):
-        normalized = f"{normalized}/v1"
-    return normalized
-
-
 def build_runpod_endpoint(pod_id: str, port: int = 11434) -> str:
-    return normalize_openai_base_url(f"https://{pod_id}-{port}.proxy.runpod.net")
-
-
-def resolve_base_url(*, vertex_base_url: str | None = None, base_url: str | None = None, pod_id: str | None = None) -> tuple[str | None, str | None]:
-    if vertex_base_url:
-        return normalize_openai_base_url(vertex_base_url), "vertex_base_url"
-    if base_url:
-        return normalize_openai_base_url(base_url), "base_url"
-    if pod_id:
-        return build_runpod_endpoint(pod_id), "pod_id"
-    return None, None
+    return f"https://{pod_id}-{port}.proxy.runpod.net/v1"
 
 
 def parse_deploy_response(payload: dict[str, Any]) -> dict[str, str]:
@@ -129,7 +102,7 @@ def update_config_text(config_text: str, *, base_url: str, model: str = DEFAULT_
 
     replacement = {
         "name": provider_name,
-        "base_url": normalize_openai_base_url(base_url),
+        "base_url": base_url,
         "api_key": "",
         "model": model,
     }
@@ -156,8 +129,7 @@ def write_config_file(config_path: Path, *, base_url: str, model: str = DEFAULT_
     return updated
 
 
-def verify_openai_chat(base_url: str, *, model: str = DEFAULT_MODEL, prompt: str = DEFAULT_VERIFY_PROMPT) -> str:
-    base_url = normalize_openai_base_url(base_url)
+def verify_openai_chat(base_url: str, *, model: str = DEFAULT_MODEL, prompt: str = "Say READY") -> str:
     payload = json.dumps(
         {
             "model": model,
@@ -167,7 +139,7 @@ def verify_openai_chat(base_url: str, *, model: str = DEFAULT_MODEL, prompt: str
         }
     ).encode()
     req = request.Request(
-        f"{base_url}/chat/completions",
+        f"{base_url.rstrip('/')}/chat/completions",
         data=payload,
         headers={"Content-Type": "application/json"},
         method="POST",
@@ -177,30 +149,6 @@ def verify_openai_chat(base_url: str, *, model: str = DEFAULT_MODEL, prompt: str
     return data["choices"][0]["message"]["content"]
 
 
-def build_vps_verify_command(
-    *,
-    base_url: str,
-    model: str = DEFAULT_MODEL,
-    prompt: str = DEFAULT_VERIFY_PROMPT,
-    vps_host: str = DEFAULT_BEZALEL_VPS_HOST,
-) -> str:
-    payload = json.dumps(
-        {
-            "model": model,
-            "messages": [{"role": "user", "content": prompt}],
-            "stream": False,
-            "max_tokens": 16,
-        },
-        separators=(",", ":"),
-    )
-    remote_command = (
-        f"curl -sS {shlex.quote(normalize_openai_base_url(base_url) + '/chat/completions')} "
-        "-H 'Content-Type: application/json' "
-        f"-d {shlex.quote(payload)}"
-    )
-    return f"ssh root@{vps_host} {shlex.quote(remote_command)}"
-
-
 def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser(description="Provision a RunPod Gemma 4 endpoint and wire a Hermes config for Bezalel.")
     parser.add_argument("--pod-name", default="bezalel-gemma4")
@@ -212,8 +160,6 @@ def parse_args() -> argparse.Namespace:
     parser.add_argument("--config-path", type=Path, default=DEFAULT_CONFIG_PATH)
     parser.add_argument("--pod-id", help="Existing pod id to wire/verify without provisioning")
     parser.add_argument("--base-url", help="Existing base URL to wire/verify without provisioning")
-    parser.add_argument("--vertex-base-url", help="OpenAI-compatible Vertex bridge URL; takes precedence over --base-url and --pod-id")
-    parser.add_argument("--vps-host", default=DEFAULT_BEZALEL_VPS_HOST, help="Bezalel VPS host for the remote curl proof command")
     parser.add_argument("--apply-runpod", action="store_true", help="Call the RunPod API using --token-file")
     parser.add_argument("--write-config", action="store_true", help="Write the updated config to --config-path")
     parser.add_argument("--verify-chat", action="store_true", help="Call the OpenAI-compatible chat endpoint")
@@ -229,18 +175,13 @@ def main() -> None:
         "cloud_type": args.cloud_type,
         "model": args.model,
         "provider_name": args.provider_name,
-        "config_path": str(args.config_path),
-        "vps_host": args.vps_host,
         "actions": [],
     }
 
-    base_url, base_url_source = resolve_base_url(
-        vertex_base_url=args.vertex_base_url,
-        base_url=args.base_url,
-        pod_id=args.pod_id,
-    )
-    if base_url_source:
-        summary["actions"].append(f"resolved_base_url_from_{base_url_source}")
+    base_url = args.base_url
+    if not base_url and args.pod_id:
+        base_url = build_runpod_endpoint(args.pod_id)
+        summary["actions"].append("computed_base_url_from_pod_id")
 
     if args.apply_runpod:
         if not args.token_file.exists():
@@ -255,17 +196,12 @@ def main() -> None:
         base_url = build_runpod_endpoint("<pod-id>")
         summary["actions"].append("using_placeholder_base_url")
 
-    summary["base_url"] = normalize_openai_base_url(base_url)
+    summary["base_url"] = base_url
     summary["config_preview"] = update_config_text("", base_url=base_url, model=args.model, provider_name=args.provider_name)
-    summary["vps_verify_command"] = build_vps_verify_command(
-        base_url=base_url,
-        model=args.model,
-        prompt=DEFAULT_VERIFY_PROMPT,
-        vps_host=args.vps_host,
-    )
 
     if args.write_config:
         write_config_file(args.config_path, base_url=base_url, model=args.model, provider_name=args.provider_name)
+        summary["config_path"] = str(args.config_path)
         summary["actions"].append("wrote_config")
 
     if args.verify_chat:
@@ -278,10 +214,8 @@ def main() -> None:
 
     print("--- Bezalel Gemma4 RunPod Wiring ---")
     print(f"Pod name: {args.pod_name}")
-    print(f"Base URL: {summary['base_url']}")
+    print(f"Base URL: {base_url}")
     print(f"Model: {args.model}")
-    print(f"Config target: {args.config_path}")
-    print(f"Bezalel VPS proof: {summary['vps_verify_command']}")
     if args.write_config:
         print(f"Config written: {args.config_path}")
     if "verify_response" in summary:

scripts/fleet_cost_report.py

@@ -0,0 +1,245 @@
+#!/usr/bin/env python3
+"""Fleet cost report generator.
+
+Reads Timmy's sovereignty metrics database and estimates paid API spend by
+agent/provider lane. Default output targets the local timmy-config reports
+folder so the cost report can be filed from the sidecar repo.
+"""
+
+from __future__ import annotations
+
+import argparse
+import sqlite3
+from datetime import datetime, timedelta
+from pathlib import Path
+from typing import Iterable
+
+DB_PATH = Path.home() / ".timmy" / "metrics" / "model_metrics.db"
+
+
+AGENT_LANES = (
+    {
+        "agent": "Timmy Cloud Lane",
+        "provider": "OpenRouter",
+        "patterns": ("openrouter/", "google/", "deepseek/", "x-ai/", "mistral/"),
+        "notes": "Cloud fallback and external reasoning routed through OpenRouter-compatible lanes.",
+    },
+    {
+        "agent": "Ezra",
+        "provider": "Anthropic",
+        "patterns": ("claude-", "anthropic/claude"),
+        "notes": "Archivist / long-form reasoning house on Claude-family models.",
+    },
+    {
+        "agent": "Bezalel",
+        "provider": "OpenAI",
+        "patterns": ("gpt-", "openai/", "codex"),
+        "notes": "Forge / implementation house on Codex/OpenAI-backed execution lanes.",
+    },
+    {
+        "agent": "Allegro",
+        "provider": "Kimi / Moonshot",
+        "patterns": ("kimi", "moonshot"),
+        "notes": "Tempo-and-dispatch house on Kimi / Moonshot direct API lanes.",
+    },
+)
+
+
+def default_report_path(report_date: str | None = None) -> Path:
+    if report_date is None:
+        report_date = datetime.now().strftime("%Y-%m-%d")
+    return Path.home() / "code" / "timmy-config" / "reports" / "production" / f"{report_date}-fleet-cost-report.md"
+
+
+def match_lane(model: str) -> dict | None:
+    lowered = (model or "").lower()
+    for lane in AGENT_LANES:
+        if any(pattern in lowered for pattern in lane["patterns"]):
+            return lane
+    return None
+
+
+def load_cost_rows(days: int = 30, db_path: Path = DB_PATH) -> list[tuple[str, int, int, int, float]]:
+    if not db_path.exists():
+        return []
+    cutoff = (datetime.now() - timedelta(days=days)).timestamp()
+    with sqlite3.connect(str(db_path)) as conn:
+        rows = conn.execute(
+            """
+            SELECT model, SUM(sessions), SUM(messages), SUM(tool_calls), SUM(est_cost_usd)
+            FROM session_stats
+            WHERE timestamp > ? AND is_local = 0
+            GROUP BY model
+            ORDER BY SUM(est_cost_usd) DESC, model ASC
+            """,
+            (cutoff,),
+        ).fetchall()
+    return [
+        (model, int(sessions or 0), int(messages or 0), int(tool_calls or 0), float(cost or 0.0))
+        for model, sessions, messages, tool_calls, cost in rows
+    ]
+
+
+def summarize_rows(rows: Iterable[tuple[str, int, int, int, float]], days: int = 30) -> dict:
+    rows = list(rows)
+    agents: dict[str, dict] = {}
+    providers_seen: set[str] = set()
+    inventory = [
+        {
+            "agent": lane["agent"],
+            "provider": lane["provider"],
+            "notes": lane["notes"],
+        }
+        for lane in AGENT_LANES
+    ]
+
+    for lane in AGENT_LANES:
+        agents[lane["agent"]] = {
+            "provider": lane["provider"],
+            "models": [],
+            "sessions": 0,
+            "messages": 0,
+            "tool_calls": 0,
+            "monthly_cost_usd": 0.0,
+            "daily_cost_usd": 0.0,
+            "notes": lane["notes"],
+        }
+
+    unassigned = {
+        "provider": "Unassigned",
+        "models": [],
+        "sessions": 0,
+        "messages": 0,
+        "tool_calls": 0,
+        "monthly_cost_usd": 0.0,
+        "daily_cost_usd": 0.0,
+        "notes": "Observed paid-model spend not yet mapped to a named wizard house.",
+    }
+
+    for model, sessions, messages, tool_calls, monthly_cost in rows:
+        lane = match_lane(model)
+        if lane is None:
+            bucket = unassigned
+        else:
+            bucket = agents[lane["agent"]]
+            providers_seen.add(lane["provider"])
+        bucket["models"].append(
+            {
+                "model": model,
+                "sessions": sessions,
+                "messages": messages,
+                "tool_calls": tool_calls,
+                "monthly_cost_usd": round(monthly_cost, 4),
+            }
+        )
+        bucket["sessions"] += sessions
+        bucket["messages"] += messages
+        bucket["tool_calls"] += tool_calls
+        bucket["monthly_cost_usd"] += monthly_cost
+
+    for bucket in list(agents.values()) + [unassigned]:
+        bucket["monthly_cost_usd"] = round(bucket["monthly_cost_usd"], 4)
+        bucket["daily_cost_usd"] = round(bucket["monthly_cost_usd"] / max(days, 1), 4)
+
+    if unassigned["models"]:
+        agents["Unassigned"] = unassigned
+        providers_seen.add("Unassigned")
+
+    total_monthly = round(sum(item["monthly_cost_usd"] for item in agents.values()), 4)
+    total_daily = round(sum(item["daily_cost_usd"] for item in agents.values()), 4)
+
+    provider_order = sorted(providers_seen)
+    if "Unassigned" in provider_order:
+        provider_order = [p for p in provider_order if p != "Unassigned"] + ["Unassigned"]
+
+    return {
+        "days": days,
+        "providers": provider_order,
+        "inventory": inventory,
+        "agents": agents,
+        "total_monthly_cost_usd": total_monthly,
+        "total_daily_cost_usd": total_daily,
+    }
+
+
+def render_markdown(summary: dict, report_date: str | None = None) -> str:
+    if report_date is None:
+        report_date = datetime.now().strftime("%Y-%m-%d")
+    lines = [
+        f"# Fleet Cost Report — {report_date}",
+        "",
+        f"Window: last {summary['days']} days of paid-model session stats from `~/.timmy/metrics/model_metrics.db`.",
+        "",
+        "## Paid API inventory",
+        "",
+        "| Agent | Provider | Notes |",
+        "| --- | --- | --- |",
+    ]
+    for item in summary["inventory"]:
+        lines.append(f"| {item['agent']} | {item['provider']} | {item['notes']} |")
+
+    lines.extend(
+        [
+            "",
+            "## Estimated cost per agent per day",
+            "",
+            "| Agent | Provider | Daily cost | Monthly estimate | Sessions | Messages | Tool calls |",
+            "| --- | --- | ---: | ---: | ---: | ---: | ---: |",
+        ]
+    )
+    for agent, data in summary["agents"].items():
+        lines.append(
+            f"| {agent} | {data['provider']} | ${data['daily_cost_usd']:.2f} | ${data['monthly_cost_usd']:.2f} | {data['sessions']} | {data['messages']} | {data['tool_calls']} |"
+        )
+
+    lines.extend(
+        [
+            "",
+            f"Total estimated daily paid spend: ${summary['total_daily_cost_usd']:.2f}",
+            f"Total estimated monthly paid spend: ${summary['total_monthly_cost_usd']:.2f}",
+            "",
+            "## Model evidence",
+            "",
+        ]
+    )
+    for agent, data in summary["agents"].items():
+        lines.append(f"### {agent}")
+        if not data["models"]:
+            lines.append("- No paid-model sessions observed in the selected window.")
+        else:
+            for model in data["models"]:
+                lines.append(
+                    f"- `{model['model']}` — {model['sessions']} sessions / {model['messages']} messages / {model['tool_calls']} tool calls / ${model['monthly_cost_usd']:.2f} est."
+                )
+        lines.append("")
+
+    lines.append("Generated by `python3 scripts/fleet_cost_report.py --days 30`. Default output path targets the local timmy-config report lane.")
+    lines.append("")
+    return "\n".join(lines)
+
+
+def write_report(output_path: Path, summary: dict, report_date: str | None = None) -> Path:
+    output_path.parent.mkdir(parents=True, exist_ok=True)
+    output_path.write_text(render_markdown(summary, report_date=report_date), encoding="utf-8")
+    return output_path
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(description="Estimate paid API spend per fleet agent")
+    parser.add_argument("--days", type=int, default=30, help="Lookback window in days")
+    parser.add_argument("--db-path", default=str(DB_PATH), help="Path to model_metrics.db")
+    parser.add_argument("--output", help="Optional markdown output path")
+    parser.add_argument("--date", help="Override report date (YYYY-MM-DD)")
+    args = parser.parse_args()
+
+    rows = load_cost_rows(days=args.days, db_path=Path(args.db_path).expanduser())
+    summary = summarize_rows(rows, days=args.days)
+    report_date = args.date or datetime.now().strftime("%Y-%m-%d")
+    output_path = Path(args.output).expanduser() if args.output else default_report_path(report_date)
+    write_report(output_path, summary, report_date=report_date)
+    print(output_path)
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

tests/test_bezalel_gemma4_vps.py

@@ -1,20 +1,14 @@
 from __future__ import annotations
 
 import json
-from pathlib import Path
 from unittest.mock import patch
 
 import yaml
 
 from scripts.bezalel_gemma4_vps import (
-    DEFAULT_CONFIG_PATH,
-    DEFAULT_BEZALEL_VPS_HOST,
     build_deploy_mutation,
     build_runpod_endpoint,
-    build_vps_verify_command,
-    normalize_openai_base_url,
     parse_deploy_response,
-    resolve_base_url,
     update_config_text,
     verify_openai_chat,
 )
@@ -34,10 +28,6 @@ class _FakeResponse:
         return False
 
 
-def test_default_config_path_targets_bezalel_vps_root_config() -> None:
-    assert DEFAULT_CONFIG_PATH == Path("/root/wizards/bezalel/home/config.yaml")
-
-
 def test_build_deploy_mutation_uses_ollama_image_and_openai_port() -> None:
     query = build_deploy_mutation(name="bezalel-gemma4", gpu_type="NVIDIA L40S", model_tag="gemma4:latest")
 
@@ -47,30 +37,6 @@ def test_build_deploy_mutation_uses_ollama_image_and_openai_port() -> None:
     assert 'volumeMountPath: "/root/.ollama"' in query
 
 
-def test_normalize_openai_base_url_adds_v1_suffix() -> None:
-    assert normalize_openai_base_url("https://pod-11434.proxy.runpod.net") == "https://pod-11434.proxy.runpod.net/v1"
-
-
-def test_normalize_openai_base_url_trims_chat_completions_suffix() -> None:
-    assert normalize_openai_base_url("https://pod-11434.proxy.runpod.net/v1/chat/completions") == "https://pod-11434.proxy.runpod.net/v1"
-
-
-def test_resolve_base_url_prefers_vertex_over_base_and_pod_id() -> None:
-    base_url, source = resolve_base_url(
-        vertex_base_url="https://vertex.example.com/openai",
-        base_url="https://plain.example.com",
-        pod_id="abc123",
-    )
-    assert source == "vertex_base_url"
-    assert base_url == "https://vertex.example.com/openai/v1"
-
-
-def test_resolve_base_url_falls_back_to_base_url_before_pod_id() -> None:
-    base_url, source = resolve_base_url(base_url="https://plain.example.com", pod_id="abc123")
-    assert source == "base_url"
-    assert base_url == "https://plain.example.com/v1"
-
-
 def test_build_runpod_endpoint_appends_v1_suffix() -> None:
     assert build_runpod_endpoint("abc123") == "https://abc123-11434.proxy.runpod.net/v1"
 
@@ -94,7 +60,7 @@ def test_parse_deploy_response_extracts_pod_id_and_endpoint() -> None:
     }
 
 
-def test_update_config_text_upserts_big_brain_provider_and_normalizes_base_url() -> None:
+def test_update_config_text_upserts_big_brain_provider() -> None:
     original = """
 model:
   default: kimi-k2.5
@@ -106,7 +72,7 @@ custom_providers:
     model: gemma3:27b
 """
 
-    updated = update_config_text(original, base_url="https://new-pod-11434.proxy.runpod.net", model="gemma4:latest")
+    updated = update_config_text(original, base_url="https://new-pod-11434.proxy.runpod.net/v1", model="gemma4:latest")
     parsed = yaml.safe_load(updated)
 
     assert parsed["model"] == {"default": "kimi-k2.5", "provider": "kimi-coding"}
@@ -120,14 +86,7 @@ custom_providers:
     ]
 
 
-def test_build_vps_verify_command_targets_bezalel_host_and_chat_completions() -> None:
-    command = build_vps_verify_command(base_url="https://pod-11434.proxy.runpod.net", model="gemma4:latest")
-    assert command.startswith(f"ssh root@{DEFAULT_BEZALEL_VPS_HOST} ")
-    assert "/v1/chat/completions" in command
-    assert "gemma4:latest" in command
-
-
-def test_verify_openai_chat_calls_chat_completions_with_normalized_base_url() -> None:
+def test_verify_openai_chat_calls_chat_completions() -> None:
     response_payload = {
         "choices": [
             {
@@ -142,7 +101,7 @@ def test_verify_openai_chat_calls_chat_completions_with_normalized_base_url() ->
         "scripts.bezalel_gemma4_vps.request.urlopen",
         return_value=_FakeResponse(response_payload),
     ) as mocked:
-        result = verify_openai_chat("https://pod-11434.proxy.runpod.net", model="gemma4:latest", prompt="say READY")
+        result = verify_openai_chat("https://pod-11434.proxy.runpod.net/v1", model="gemma4:latest", prompt="say READY")
 
     assert result == "READY"
     req = mocked.call_args.args[0]
@@ -150,10 +109,3 @@ def test_verify_openai_chat_calls_chat_completions_with_normalized_base_url() ->
     payload = json.loads(req.data.decode())
     assert payload["model"] == "gemma4:latest"
     assert payload["messages"][0]["content"] == "say READY"
-
-
-def test_readme_documents_root_config_path_and_vps_proof_command() -> None:
-    readme = Path("scripts/README_bezalel_gemma4_vps.md").read_text()
-    assert "/root/wizards/bezalel/home/config.yaml" in readme
-    assert "ssh root@104.131.15.18" in readme
-    assert "--vertex-base-url" in readme

tests/test_fleet_cost_report.py

@@ -0,0 +1,77 @@
+from importlib.util import module_from_spec, spec_from_file_location
+from pathlib import Path
+import tempfile
+import unittest
+
+
+ROOT = Path(__file__).resolve().parent.parent
+SCRIPT_PATH = ROOT / "scripts" / "fleet_cost_report.py"
+
+
+def load_module():
+    spec = spec_from_file_location("fleet_cost_report", SCRIPT_PATH)
+    module = module_from_spec(spec)
+    assert spec.loader is not None
+    spec.loader.exec_module(module)
+    return module
+
+
+class TestFleetCostReport(unittest.TestCase):
+    def test_default_output_targets_timmy_config_report_path(self):
+        module = load_module()
+        output_path = module.default_report_path("2026-04-22")
+        self.assertIn("timmy-config", str(output_path))
+        self.assertTrue(str(output_path).endswith("2026-04-22-fleet-cost-report.md"))
+
+    def test_summary_groups_paid_costs_by_agent_and_provider(self):
+        module = load_module()
+        rows = [
+            ("claude-sonnet-4-6", 12, 120, 24, 6.0),
+            ("gpt-5.4", 6, 60, 12, 3.0),
+            ("openrouter/google/gemini-2.5-pro", 4, 40, 8, 2.0),
+            ("kimi-k2", 2, 20, 4, 1.0),
+        ]
+        summary = module.summarize_rows(rows, days=30)
+
+        self.assertEqual(summary["providers"], ["Anthropic", "Kimi / Moonshot", "OpenAI", "OpenRouter"])
+        self.assertAlmostEqual(summary["agents"]["Ezra"]["monthly_cost_usd"], 6.0)
+        self.assertAlmostEqual(summary["agents"]["Bezalel"]["monthly_cost_usd"], 3.0)
+        self.assertAlmostEqual(summary["agents"]["Timmy Cloud Lane"]["monthly_cost_usd"], 2.0)
+        self.assertAlmostEqual(summary["agents"]["Allegro"]["monthly_cost_usd"], 1.0)
+        self.assertAlmostEqual(summary["agents"]["Ezra"]["daily_cost_usd"], 0.2)
+
+    def test_report_render_mentions_inventory_and_agent_costs(self):
+        module = load_module()
+        rows = [
+            ("claude-sonnet-4-6", 12, 120, 24, 6.0),
+            ("gpt-5.4", 6, 60, 12, 3.0),
+            ("openrouter/google/gemini-2.5-pro", 4, 40, 8, 2.0),
+        ]
+        summary = module.summarize_rows(rows, days=30)
+        report = module.render_markdown(summary, report_date="2026-04-22")
+
+        self.assertIn("# Fleet Cost Report — 2026-04-22", report)
+        self.assertIn("## Paid API inventory", report)
+        self.assertIn("Anthropic", report)
+        self.assertIn("OpenRouter", report)
+        self.assertIn("OpenAI", report)
+        self.assertIn("## Estimated cost per agent per day", report)
+        self.assertIn("Timmy Cloud Lane", report)
+        self.assertIn("Ezra", report)
+        self.assertIn("Bezalel", report)
+
+    def test_write_report_creates_markdown_file(self):
+        module = load_module()
+        rows = [("claude-sonnet-4-6", 1, 10, 2, 0.5)]
+        summary = module.summarize_rows(rows, days=30)
+        with tempfile.TemporaryDirectory() as tmpdir:
+            dest = Path(tmpdir) / "fleet-cost.md"
+            module.write_report(dest, summary, report_date="2026-04-22")
+            self.assertTrue(dest.exists())
+            text = dest.read_text()
+            self.assertIn("Fleet Cost Report", text)
+            self.assertIn("Ezra", text)
+
+
+if __name__ == "__main__":
+    unittest.main()