ai-ledger

Practical Python, AI, and automation for accounting and finance teams. Built for professionals who want to work faster and go deeper — without becoming developers.

Each article includes the full write-up, sample data, and the code to reproduce every chart.


Articles

# Title Topics
00 True or False: AI Will Replace My Job? Career perspective, Python fundamentals, durable skills, why Python Muse exists
00.5 Where to Start If You’re Ready to Work With AI Getting started, VS Code, Python setup, Microsoft Copilot, mindset shift
01 Your AI Co-Pilot for Accounting Margin analysis, vendor cost inflation, salesperson performance, revenue concentration
02 Ways to Use Claude: Choosing the Right Interface Claude.ai, API, Claude Code, interface comparison, getting started
03 Getting the Right Tools Installed: A Safe Starting Point for Accounting Teams IT approvals, secure adoption, VS Code and Python setup framing, low-risk pilot use cases
04 AI in Accounting Is Not the Wild West Anymore AI governance, COSO framework, PCAOB updates, Big Four guidance, audit-ready workflows
05 Reproducible Accounting Accounting as code, audit evidence, version control, reproducible financial reporting
06 How to Use AI in Accounting Without Sending the Wrong Data Data masking, local vs cloud processing, safe AI workflows, QuickBooks demo, validation hooks
07 AI Governance for Controllers COSO framework, controller governance, use case inventory, AI skills, agents, VS Code
08 Why Claude “Forgets” – And How to Fix It Context window, external memory, project files, plan.md, status_update.md, CLAUDE.md
09 How Accountants Learn AI Excel-to-AI learning path, 13-skill framework, Markdown, Python, Git, hooks, canary, project hygiene
10 AI in Accounting: Real Use Cases – and How to Structure Them Reconciliations, variance analysis, ad hoc analysis, 3-tier classification, exploratory vs repeatable vs audit-ready
11 From One-Time Analysis to Repeatable Workflows 9-step workflow pattern, data masking, headers-only, hooks, plan approval, SKILL documentation
12 When to Trust AI to Run Your Accounting Workflows Audit-ready framework, COSO mapping, governance controls, logged execution, canary checks
13 AI in Accounting Isn’t Just About Efficiency – It’s About Control Zero Trust for AI, OWASP LLM risks, data controls, trust but verify, four-level framework, checklist
14 Stop Using One AI Like It Is Excel Claude vs ChatGPT vs Copilot, model orchestration, skills, GitHub as distribution layer, folder structure as prompt
15 “AI Can’t Work With Our Excel Files”… or Can It? Excel instruction layer, SKILL for spreadsheets, three-tier data approach, train once reuse forever
16 The PDF Token Trap PDF-to-Markdown workflow, token efficiency, data masking, reusable Skills, convert once reuse forever
17 The Power of Skills and Agents: How Accountants Actually Use AI Skills, agents, workflow design, internal controls, living documentation
17b Your First CLAUDE.md: How Accountants Define the Agent CLAUDE.md structure, governance documentation, accounting-specific rules, workflow kit
18 Scheduled Workflows for Accounting Teams Scheduled tasks, cron, automation, recurring workflows
19 Multi-Agent Orchestration for Accountants Multi-agent, orchestration, parallel workflows, accounting automation
20 Version Control for Accountants Git, version control, audit trail, reproducibility
21 The 3 Mindsets of AI Adoption in Accounting Mindset, change management, Builder/Stabilizer/Protector, AI adoption
22 The Workings Layer: Fitting AI Into the Files You Can’t Change Workings layer, folder structure, CLAUDE.md placement, legacy files, audit-ready
23 Don’t Trust the Model to Find What You Already Know Is There Schema-driven sanitization, PII detection, data privacy, accounting data workflows, Privacy Filter
24 AI Didn’t Break Your Numbers. Excel Did. Excel formatting, accounting number normalization, data discipline, Top 10 AI Traps #1
25 What the Heck Is a Script? Scripts explained for accountants, Excel vs Python readability, SOPs for computers, audit-friendly logic
26 When Your AI Enters Month-End Close Mode Context drift, session management, context window, accounting analogies, SKILL files, checkpoints
27 Visual Studio Code Extensions for Accountants VS Code extensions, governance, security, minimalism, extensions vs libraries, starter stack
28 Python Libraries for Accountants: Skills You Teach Your Code Python libraries, pandas, openpyxl, matplotlib, pip install, environments, extensions vs libraries
29 The Magic Loop: Why Easy to Generate Doesn’t Mean Safe to Run Workflow automation, YAML configuration, loop governance, review before run, abstraction layers
30 AI Routines for Accountants: When Your Guidance Starts Checking Itself AI routines, monitoring workflows, proposed change packages, human review gates, audit evidence, tax guidance example
31 Metadata Is the Label Maker Your AI Workflow Needs Metadata governance, SKILL.md, file manifests, scripts vs. controls, hooks, audit trail
32 From AI Answers to Audit Trails: How Accountants Can Validate AI Output AI output validation, chat vs. harness, tie-out reports, evidence trails, reviewer checklists

How to Use This Repo

Each article folder is self-contained:

articles/
└── 01-ai-copilot-for-accounting/
    ├── README.md            ← the article
    ├── visuals/             ← charts referenced in the article
    ├── data/                ← sample CSV files
    └── generate_visuals.py  ← script to reproduce all charts

To reproduce the analysis in any article:

  1. Download the data/ folder
  2. Open it in VS Code with Claude Code installed
  3. Follow the prompts in the article

Who This Is For

Contributions are welcome from:

Whether you write code or review accounting logic, there’s a way to help.


Current Contribution Needs

See the full list at community/ideas-wanted.md.


How to Contribute

Open an issue, comment on an existing request, or submit a pull request.

See CONTRIBUTING.md for details and community/roadmap.md for where the project is headed.


License

This project uses a dual license:

See LICENSE and LICENSE-CODE for full details.


⚠️ Disclaimer

This repository is provided for educational purposes only.


AI Usage Notice

This repository may include AI-assisted workflows and scripts.

Users must:

PythonMuse LLC is not responsible for any outcomes resulting from use of this code.


By Svetlana Toohey