gabrielserrao/claude-for-reservoir-engineering

★ 22PythonAudience · researcherComplexity · 2/5Setup · moderate

Mindmap

mindmap
  root((repo))
    Nine exercises
      Production data quality
      Decline curve analysis
      PVT calculations
      Parallel sensitivity runs
    Key idea
      Engineer stays in control
      Specific prompts win
      Tests with known values
    Tech stack
      Python 3.10+
      Claude Code CLI
      pyResToolbox
    Audience
      Reservoir engineers
      Petroleum engineers

mindmap root((repo)) Nine exercises Production data quality Decline curve analysis PVT calculations Parallel sensitivity runs Key idea Engineer stays in control Specific prompts win Tests with known values Tech stack Python 3.10+ Claude Code CLI pyResToolbox Audience Reservoir engineers Petroleum engineers

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Work through nine hands-on exercises to learn how to write Claude Code prompts that produce reliable reservoir engineering calculations instead of guessed formulas.

USE CASE 2

Integrate pyResToolbox as a tool server so Claude Code runs peer-reviewed PVT and decline curve functions rather than generating its own equations.

USE CASE 3

Set up a reviewer agent inside Claude Code that automatically flags unit and correlation mistakes in petroleum engineering scripts before they reach production.

Tech stack

PythonClaude Code

Getting it running

Difficulty · moderate Time to first run · 30min

Requires Python 3.10+, the Claude Code CLI, and optionally pyResToolbox for the tool server exercises.

No license information is mentioned in the explanation.

In plain English

This repository is a training course that teaches reservoir and petroleum engineers how to use Claude Code, an AI-assisted coding tool, for real engineering work. It is written for engineers who already understand domain concepts like production data, decline curve analysis, and PVT calculations, but who want to use AI to speed up their coding without letting the AI make unchecked assumptions. The course is organized into nine exercises, each focused on a specific engineering workflow: reviewing production data quality, analyzing water-cut trends, writing decline curve sanity checks, creating persistent project memory files, packaging reusable workflows as skills, adding a reviewer agent that catches unit and correlation mistakes, running shell-based production CSV checks, calling peer-reviewed petroleum engineering functions through an AI-compatible tool server, and running parallel sensitivity studies. Each exercise shows a vague prompt first, then a better one, so the reader can compare what changes and why. The central idea is that the engineer stays in control. Claude Code produces more dependable technical work when the person directing it names the specific file, explains the units, states the expected physical behavior, and requests tests with known reference values. The course demonstrates this pattern through concrete, repeatable examples rather than abstract guidance. The technical setup requires Python 3.10 or newer and the Claude Code command-line tool. An optional library called pyResToolbox provides peer-reviewed petroleum engineering calculations, and a companion tool server exposes those calculations to Claude Code during live sessions. This lets engineers ask Claude to run real PVT calculations through a tested library rather than having it produce its own formulas from scratch. The course was adapted from a similar project built for hydrology, with every exercise rewritten around reservoir engineering problems that practicing engineers will recognize from their daily work.

Copy-paste prompts

Prompt 1

I am on Exercise 2 of claude-for-reservoir-engineering covering water-cut trend analysis. Write a prompt that tells Claude Code the exact CSV file path, column names for water cut and date, and the expected physical trend so it produces a trustworthy analysis.

Prompt 2

Using the approach from claude-for-reservoir-engineering, help me create a CLAUDE.md project memory file that stores the well name, reservoir units, and reference decline rate so I do not have to repeat them in every prompt.

Prompt 3

I want to connect pyResToolbox to Claude Code as a tool server for live PVT calculations. Walk me through the setup steps from the claude-for-reservoir-engineering course.

Prompt 4

Following the reviewer agent pattern from claude-for-reservoir-engineering, write a system prompt for an agent that automatically checks reservoir engineering scripts for unit inconsistencies between psi, bar, and kPa.

Open on GitHub → Explain another repo

← gabrielserrao on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.