digital-innovation-agents

Digital Innovation Agents

When code costs almost nothing, the plan becomes the product.

A V-Model workflow that walks your AI coding assistant through Business Analysis, Requirements Engineering, Architecture, Coding, Testing, and a Security Audit, with quality-gated handoffs between every phase.

Full documentation: pssah4.github.io/digital-innovation-agents

Shipping code is a solved problem. What most teams still lack is evidence that the features they ship matter to a real user. Digital Innovation Agents pair a battle-tested innovation methodology with a state-of-the-art coding workflow, so your AI never builds the wrong thing at speed. Works across Claude Code, Cursor, Codex, OpenCode, Gemini CLI, and GitHub Copilot.

What this is

The project ships thirteen skills that run inside your AI coding assistant. Six are V-Model phase skills (business analysis, requirements engineering, architecture, coding, testing, security audit). Two are entry-point skills for non-greenfield projects (reverse engineering, dia-migration). One is the on-demand workflow guide (/dia-guide). Four are foundation skills (project conventions, consistency check, humanizer, dia-bootstrap). Every phase skill owns one part of the V-Model, has its own quality gates, and hands off a structured artifact to the next phase. The guide is called separately whenever the user wants an orientation read. Every decision stays traceable from a real user problem through requirements, architecture, code, tests, and a security audit.

Three entry points cover greenfield, brownfield, and migration projects:

• Greenfield: /business-analysis starts with structured discovery (users, needs, insights, critical hypotheses) and walks forward through the V-Model.
• Brownfield: /reverse-engineering walks the V backwards over an existing codebase and produces plan-context, ADRs, an arc42 snapshot, a FEATURE inventory, a backlog seed, and an evidence-based BA draft. Every claim is sourced to a file path or doc section. Nothing invented.
• Migration: /dia-migration brings an older DIA project (v1) or a pre-existing V-Model variant up to current conventions: cleans status drift, normalises ID schemas, flattens analysis/, regenerates the backlog as single source of truth.

Innovation methodology, not just automation

The BA and RE agents ship a catalog of 32 innovation methods (qualitative interviews, extreme users, fly on the wall, cultural probes, persona synthesis, stakeholder maps, jobs to be done, brainwriting, TRIZ, wizard of oz, pre-mortem, value proposition quantification, and more) organised as method cards in the docs.

During a BA or RE session, when your answers go thin, the agent stops the interview and proposes the matching field method with a one-page card: what it produces, when to reach for it, how to run it, team and time, things that go wrong, and what to bring back to the session. The actual research work stays human-to-human: interviews with real users, observations in the real context, prototypes on real hands. The agent's job is to spot the gap and pick the right method, not to replace the work.

Quick start

Pick your platform. Each installer drops the same thirteen skills with the same templates and quality gates into your tool of choice.

Claude Code (recommended)

/plugin lives in the Claude Code CLI, not in the VS Code or JetBrains extensions. If claude --version returns command not found, install the CLI first:

curl -fsSL https://claude.ai/install.sh | bash    # official installer
# or: brew install --cask claude-code              # macOS Homebrew
# or: npm install -g @anthropic-ai/claude-code     # any OS with Node

Reopen your shell (source ~/.zshrc or ~/.bashrc), then install the plugin:

claude

/plugin marketplace add https://github.com/pssah4/digital-innovation-agents.git
/plugin install digital-innovation-agents@pssah4-skills

Type / in any new session to see the skills in autocomplete. The dia-bootstrap skill loads automatically at session start as a brief orientation.