March 7, 2026

Summary

An internal engineering function managed a growing estate of approximately 40 distinct software repositories. As the portfolio expanded, its reliance on decentralised and inconsistent planning methods, primarily ad-hoc markdown task lists within each repository, began to hinder visibility and operational efficiency. There was no single source of truth to view ongoing work, track progress, or manage priorities at the portfolio level, creating a significant governance gap.

Challenge

The core challenge was to move from fragmented, per-repository task tracking to a unified operational model. The organisation needed a scalable and repeatable framework to standardise planning, execution, and reporting across its entire engineering estate. This required a solution that could be implemented without disrupting existing development work while providing a much-needed portfolio-level view for management and governance teams.

Objectives

To address this challenge, a project was initiated with four primary objectives:

1. Centralise Visibility: Establish a single, consolidated view of all engineering workstreams, issues, and pull requests from across the multi-repository estate.
2. Standardise Processes: Implement a consistent, issue-led planning and tracking model to replace disparate markdown-based task lists.
3. Automate Migration: Develop tooling to efficiently migrate legacy task lists and backfill existing repository issues into the new, structured system.
4. Create a Repeatable Framework: Build a documented and maintainable operating model that could be applied to both existing and future repositories with predictable results.

Approach and Delivery

The solution was an internal operations framework built around a central control repository. This repository acted as the administrative hub, holding shared configuration, automation scripts, and reusable assets. The technical approach was centred on leveraging native GitHub capabilities, specifically GitHub Projects v2, to create a ‘Master Plan’ board for portfolio-wide visibility, respecting the operational constraints of a GitHub Free environment.

The delivery was structured to support staged adoption. It included the development of Python and Bash automation for bulk operations, alongside per-repository GitHub Actions workflows that integrated individual projects into the central board. The framework was supported by comprehensive runbooks covering every stage of the process, from initial project bootstrap and field discovery to repository onboarding and legacy data import.

Technical Implementation

The framework was implemented using Python 3.9+, Bash, and GitHub Actions YAML. It interacted with the GitHub platform via the GitHub CLI and both REST and GraphQL APIs, using requests and PyYAML for core utility functions in Python.

Key components included:

• Central Projects v2 Board: A single ‘Master Plan’ board configured to aggregate issues and pull requests from all participating repositories.
• Per-Repository Workflows: GitHub Actions workflows installed in each repository to automatically add new issues and pull requests to the central project board and synchronise their status upon closure or reopening.
• Backfill and Import Tooling: A Python utility was developed to backfill existing issues from repositories into the project board, resolving all necessary metadata and status options. A separate script parsed legacy markdown task lists, created corresponding GitHub Issues via the API, and applied appropriate labels.
• Bulk Repository Configuration: A script automated the application of GitHub Actions secrets and variables across a maintained list of repositories, ensuring consistent and secure configuration for accessing the Projects v2 board.
• Standardised Templates: [PLAN] and [TASK] issue templates were created to enforce a consistent planning and work-breakdown structure across the estate.

Outcome

The project successfully established a consistent, issue-first operating model, replacing ad-hoc task tracking and providing a single portfolio view across the entire engineering landscape. The framework delivered a clear, real-time picture of work in progress, improving planning accuracy and management oversight.

The automation tooling enabled a smooth transition, allowing teams to backfill thousands of existing issues and import legacy plans without manual effort. The result was a scalable governance framework that enhanced visibility and control while operating entirely within the native GitHub toolchain.

Risks, Controls and Governance

Governance was a foundational component of the framework’s design. A key control was the separation of credentials, with documented processes for using local .env files for development and repository-level Actions secrets for production workflows. All PAT (Personal Access Token) scope requirements were explicitly defined.

The framework’s architecture, favouring explicit, per-repository workflow installation over opaque global automation, provided a critical governance control, ensuring that all changes were reviewable and deliberate. The delivery also included pre-publication checklists for sanitising secrets, maintaining clean commit histories, and reviewing dependencies, establishing a robust security posture for the operational tooling itself.

Key Lessons

1. Centralised Governance for Distributed Estates: A central control repository is an effective model for applying consistent operational standards and automation to a distributed, multi-repository environment.
2. Leverage Native Platform Capabilities: Combining native platform features like GitHub Projects with targeted automation can deliver powerful portfolio management capabilities without the complexity and cost of third-party tooling.
3. Staged Adoption is Crucial for Migration: Providing automated tooling for backfilling existing work and importing legacy data is critical for achieving buy-in and ensuring a smooth, low-friction transition to a new operating model.

Related Services

• Prompt Governance & Approval
  Prompt governance and approval services providing lifecycle management, ownership, versioning, audit trails, and controlled change for production AI prompts.
• Agentic AI & Orchestrated Workflows
  Design and deliver agentic AI workflows with multi-step orchestration, approvals, monitoring, and guardrails for controlled execution across business systems.
• PowerShell Automation & Scripting
  PowerShell automation service delivering safe tenant operations, reporting and bulk changes across Microsoft 365, Azure and endpoints with auditable scripts.