greenpick 0.6.1__tar.gz → 0.6.2__tar.gz

{greenpick-0.6.1/src/greenpick.egg-info → greenpick-0.6.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: greenpick
-Version: 0.6.1
+Version: 0.6.2
 Summary: Design-time sustainability co-pilot (MCP server): greener, cheaper cloud infrastructure decisions with carbon + cost + latency tradeoffs.
 Author-email: "Data Science Advisers, LLC" <info@datascienceadvisers.com>
 License-Expression: MIT
@@ -308,6 +308,10 @@ with open("greenpick-server.log", "w") as errlog:
 Top-level `await` works in Jupyter. In a plain script, put the `with` block inside an `async def`
 and run it with `asyncio.run(...)`.
 
+Using an AI assistant inside a notebook? Paste the
+[starter prompt](https://github.com/bithug01/greenpick/blob/master/docs/using-from-a-notebook.md)
+as your first message: it sets up the connection and points the assistant at the right tool for each question.
+
 Every carbon figure includes a `quality` field: `live` means a current reading, `typical` means
 this month's average from last year's real data, `annual_average` means a sourced yearly national
 figure (regions with no free hourly data), and `mock` means a hand-set value. Check it to confirm

{greenpick-0.6.1 → greenpick-0.6.2}/README.md

@@ -278,6 +278,10 @@ with open("greenpick-server.log", "w") as errlog:
 Top-level `await` works in Jupyter. In a plain script, put the `with` block inside an `async def`
 and run it with `asyncio.run(...)`.
 
+Using an AI assistant inside a notebook? Paste the
+[starter prompt](https://github.com/bithug01/greenpick/blob/master/docs/using-from-a-notebook.md)
+as your first message: it sets up the connection and points the assistant at the right tool for each question.
+
 Every carbon figure includes a `quality` field: `live` means a current reading, `typical` means
 this month's average from last year's real data, `annual_average` means a sourced yearly national
 figure (regions with no free hourly data), and `mock` means a hand-set value. Check it to confirm

{greenpick-0.6.1 → greenpick-0.6.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "greenpick"
-version = "0.6.1"
+version = "0.6.2"
 description = "Design-time sustainability co-pilot (MCP server): greener, cheaper cloud infrastructure decisions with carbon + cost + latency tradeoffs."
 readme = "README.md"
 requires-python = ">=3.11"

{greenpick-0.6.1 → greenpick-0.6.2}/src/greenpick/__init__.py

@@ -4,7 +4,7 @@ Greener cloud-region choices with honest carbon + cost + latency tradeoffs,
 delivered inside the vibe-coding loop. Directional guidance, not audited accounting.
 """
 
-__version__ = "0.6.1"
+__version__ = "0.6.2"
 
 # httpx logs every request URL at INFO, and our live sources pass their API key / token as a URL
 # query parameter — so at the default log level the secret would land in the MCP client's logs.

{greenpick-0.6.1 → greenpick-0.6.2}/src/greenpick/server.py

@@ -60,6 +60,18 @@ accuracy. Use explain_methodology_and_limits when asked how a number was produce
 """
 
 mcp = FastMCP("greenpick", instructions=INSTRUCTIONS)
+
+
+def _fraction_error(name: str, value: Optional[float], allow_zero: bool = False) -> Optional[str]:
+    """Utilisation inputs are fractions (0.15 = 15%). A value like 15 is almost always a percentage;
+    say so rather than silently reading it as 1,500%."""
+    if value is None:
+        return None
+    low_ok = value >= 0 if allow_zero else value > 0
+    if low_ok and value <= 1:
+        return None
+    hint = f" Did you mean {value / 100:g}?" if 1 < value <= 100 else ""
+    return (f"{name} must be a fraction between 0 and 1 (0.15 means 15%); got {value:g}.{hint}")
 _repo = get_repository()
 
 
@@ -247,6 +259,11 @@ async def rightsizing_check(
     Returns the right-sized target (keeps ~40% headroom) with the monthly $ AND kg CO2 saving. Lead
     with the cost number, carbon is the twin. Directional, design-time — uses the utilisation you
     report, not a reading of your fleet."""
+    for name, value, zero in (("cpu_utilization", cpu_utilization, False),
+                              ("ram_utilization", ram_utilization, True)):
+        err = _fraction_error(name, value, allow_zero=zero)
+        if err:
+            return {"provider": provider, "region": region, "error": err}
     advice = await rightsizing_service.rightsizing_check(
         _repo, provider, region, vcpu, ram_gb, cpu_utilization,  # type: ignore[arg-type]
         ram_utilization=ram_utilization, hours_per_month=hours_per_month,
@@ -470,7 +487,11 @@ async def estimate_training_carbon(
     portable and deferrable, so this combines region choice with cleanest-hours timing. Directional
     (TDP-based energy); GPU availability per region is NOT checked - verify your card is offered there.
     Training data is often large: if your dataset lives in another region, run data_locality_egress_check
-    before shipping the job to the greenest region - the data egress can outweigh the carbon saving."""
+    before shipping the job to the greenest region - the data egress can outweigh the carbon saving.
+    `utilization` is an optional fraction of TDP (0..1, e.g. 0.6 for 60%)."""
+    err = _fraction_error("utilization", utilization)
+    if err:
+        return {"gpu": gpu, "error": err}
     advice = await training_service.estimate_training_carbon(
         _repo, gpu, gpu_count, gpu_hours,  # type: ignore[arg-type]
         region=region, provider=provider, utilization=utilization,  # type: ignore[arg-type]

{greenpick-0.6.1 → greenpick-0.6.2/src/greenpick.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: greenpick
-Version: 0.6.1
+Version: 0.6.2
 Summary: Design-time sustainability co-pilot (MCP server): greener, cheaper cloud infrastructure decisions with carbon + cost + latency tradeoffs.
 Author-email: "Data Science Advisers, LLC" <info@datascienceadvisers.com>
 License-Expression: MIT
@@ -308,6 +308,10 @@ with open("greenpick-server.log", "w") as errlog:
 Top-level `await` works in Jupyter. In a plain script, put the `with` block inside an `async def`
 and run it with `asyncio.run(...)`.
 
+Using an AI assistant inside a notebook? Paste the
+[starter prompt](https://github.com/bithug01/greenpick/blob/master/docs/using-from-a-notebook.md)
+as your first message: it sets up the connection and points the assistant at the right tool for each question.
+
 Every carbon figure includes a `quality` field: `live` means a current reading, `typical` means
 this month's average from last year's real data, `annual_average` means a sourced yearly national
 figure (regions with no free hourly data), and `mock` means a hand-set value. Check it to confirm

{greenpick-0.6.1 → greenpick-0.6.2}/tests/test_services.py

@@ -1990,3 +1990,46 @@ def test_move_tools_point_to_the_justification_tool():
     assert "region_move_justification" in desc["compare_regions_tri_axis"]
     assert "region_move_justification" in desc["greener_alternative_to"]
     assert "FULL dollar range" in desc["migration_payoff"]
+
+
+# --- docs/using-from-a-notebook.md stays in step with the real tool list ---
+def test_notebook_prompt_lists_every_tool_and_no_unknown_ones():
+    import asyncio, re
+    from pathlib import Path
+    from greenpick import server
+    names = {t.name for t in asyncio.run(server.mcp.list_tools())}
+    text = (Path(__file__).resolve().parent.parent / "docs" / "using-from-a-notebook.md").read_text(encoding="utf-8")
+    missing = sorted(n for n in names if n not in text)
+    assert not missing, f"add these tools to docs/using-from-a-notebook.md: {missing}"
+    routed = set()
+    for line in text.splitlines():
+        if line.startswith("- ") and " -> " in line:
+            routed |= set(re.findall(r"[a-z][a-z0-9]*(?:_[a-z0-9]+)+", line.split(" -> ", 1)[1]))
+    assert routed and routed <= names, f"unknown tool names in the routing list: {sorted(routed - names)}"
+
+
+# --- 0.6.2: a percentage passed as a utilisation fraction is rejected, not read as 1,500% ---
+@pytest.mark.asyncio
+@pytest.mark.parametrize("bad", [15, 1.5, 0, -0.1, 150])
+async def test_rightsizing_rejects_non_fraction_cpu_utilization(bad):
+    r = await _tool("rightsizing_check")(provider="aws", region="us-east-1", vcpu=8, ram_gb=32, cpu_utilization=bad)
+    assert "error" in r and "fraction" in r["error"]
+    if bad == 15:
+        assert "Did you mean 0.15?" in r["error"]
+
+
+@pytest.mark.asyncio
+async def test_rightsizing_fraction_still_works_and_ram_is_checked(monkeypatch):
+    from greenpick import server
+    monkeypatch.setattr(server, "_repo", MockCarbonRepository())
+    ok = await _tool("rightsizing_check")(provider="aws", region="us-east-1", vcpu=8, ram_gb=32, cpu_utilization=0.15)
+    assert "error" not in ok and ok["recommended_vcpu"] == 4
+    bad = await _tool("rightsizing_check")(provider="aws", region="us-east-1", vcpu=8, ram_gb=32,
+                                           cpu_utilization=0.15, ram_utilization=20)
+    assert "ram_utilization" in bad["error"] and "0.2" in bad["error"]
+
+
+@pytest.mark.asyncio
+async def test_training_rejects_percentage_utilization():
+    r = await _tool("estimate_training_carbon")(gpu="a100", gpu_count=8, gpu_hours=40, utilization=60)
+    assert "error" in r and "Did you mean 0.6?" in r["error"]