PlotAI¶

AI-powered plot analysis using OpenAI. Ask questions, get ratings, and receive natural language summaries of any PlotReport.

Optional dependency — not included in the base install.
pip install plot-finder[ai]
This installs openai.

Quick Start¶

from plot_finder import Plot, PlotAnalyzer, PlotReporter
from plot_finder.ai import PlotAI

plot = Plot(plot_id="141201_1.0001.6509")
analyzer = PlotAnalyzer(plot, radius=2000)
report = PlotReporter(analyzer).report()

ai = PlotAI(report, api_key="sk-...")
print(ai.summary())

Constructor¶

PlotAI(report, *, api_key, model="gpt-4o-mini")

Parameter	Type	Default	Description
`report`	`PlotReport`	required	Report to analyze
`api_key`	`str`	required	OpenAI API key
`model`	`str`	`"gpt-4o-mini"`	OpenAI model to use

Methods¶

All methods return str.

`summary()`¶

Natural language summary of the plot and surroundings — location, nearby infrastructure, transport, nature, air quality, and sunlight.

print(ai.summary())
# "The plot 141201_1.0001.6509 is located in central Warsaw..."

`rate(purpose="living")`¶

Rate the plot 1–10 for a specific purpose with explanation.

Parameter	Type	Default	Description
`purpose`	`str`	`"living"`	Purpose to rate for

print(ai.rate())                        # living (default)
print(ai.rate(purpose="investment"))    # investment
print(ai.rate(purpose="farming"))       # farming

`advantages()`¶

Key advantages of this location based on the report data.

print(ai.advantages())
# "1. Excellent transport links — 3 bus stops within 500m..."

`disadvantages()`¶

Key disadvantages and risks of this location.

print(ai.disadvantages())
# "1. No schools within walking distance..."

`ask(question)`¶

Freeform Q&A about the plot.

Parameter	Type	Description
`question`	`str`	Any question about the plot

print(ai.ask("Is this a good place for a family with kids?"))
print(ai.ask("What are the nearest parks?"))
print(ai.ask("How is the air quality?"))

Model Selection¶

The default model is gpt-4o-mini — fast and cheap. For more detailed analysis, use a larger model:

ai = PlotAI(report, api_key="sk-...", model="gpt-4o")

Examples¶

Full analysis¶

from plot_finder import Plot, PlotAnalyzer, PlotReporter
from plot_finder.ai import PlotAI

plot = Plot(plot_id="141201_1.0001.6509")
analyzer = PlotAnalyzer(plot, radius=3000)
report = PlotReporter(analyzer).report()

ai = PlotAI(report, api_key="sk-...")

print("=== Summary ===")
print(ai.summary())

print("\n=== Rating (living) ===")
print(ai.rate())

print("\n=== Rating (investment) ===")
print(ai.rate(purpose="investment"))

print("\n=== Advantages ===")
print(ai.advantages())

print("\n=== Disadvantages ===")
print(ai.disadvantages())

print("\n=== Q&A ===")
print(ai.ask("Is this a good place for a family with kids?"))

Using environment variable for API key¶

import os
from plot_finder.ai import PlotAI

ai = PlotAI(report, api_key=os.environ["OPENAI_API_KEY"])