№ 00 AI WORKFLOW READINESS SCORECARD

How production-ready are your team's AI coding workflows?

Most teams are using AI coding tools. Few have the workflows to make them reliable. Benchmark your team in 3 minutes.

Based on 27 AI development patterns. Built by Paul Duvall, author of the Jolt Award-winning book Continuous Integration (Addison-Wesley).

Drawn from the open-source ai-development-patterns framework (400+ stars on GitHub).

15

Questions

Across 3 pillars

3 min

To Complete

Quick assessment

30-day

Action Plan

Personalized to your score

What You'll Assess

1

Context Persistence

How well your team maintains context for AI tools across sessions, contributors, and projects.

2

Multi-Agent Orchestration

How effectively you decompose work and coordinate multiple AI agents for complex tasks.

3

CI/CD Integration

How your pipeline handles AI-generated code with quality gates, testing, and deployment guardrails.

What You'll Get in Your Inbox

Your score, per-pillar analysis with specific patterns to apply, and a 30-day action plan — emailed to you as soon as you finish.

Scoring is weighted 0–100 and fully deterministic — same inputs, same score, letter grades A–F. No AI in the scoring logic.

C

44/100

Assisted

Context Persistence

C44/100

Multi-Agent Orchestration

C44/100

CI/CD Integration

C44/100

AI helps, but you're still doing the heavy lifting. Standardizing your approach will turn individual wins into team-wide leverage.

Detailed Analysis & Action Items

C Context Persistence Building Foundation — 44/100

You have some context in place, but it's likely informal or inconsistent across repos. AI tools sometimes match your conventions but often need correction. The Context Persistence pattern addresses this systematically.

Implement progressive disclosure (see Progressive Disclosure). Instead of one massive context file, create an agent_docs/ folder with topic-specific files that AI loads on demand.
Standardize context files across repos. Create a template rules file that every new project starts with. The Development Workflow rules provide a solid starting template.

C Multi-Agent Orchestration Building Foundation — 44/100

Free. No credit card. Results delivered to your inbox.

What Best Describes Your Role?

This helps us tailor your 30-day action plan to the decisions you actually make.

Select the role closest to yours:

Individual Contributor — I write code and use AI tools directly Team Lead / Engineering Manager — I guide a team's AI adoption Executive / CTO / VP — I set AI strategy for the organization

Pillar 1: Context Persistence

How well does your team maintain context for AI coding tools across sessions, contributors, and projects?

0 of 5 answered

1. Does your team use persistent context files (e.g., CLAUDE.md, rules files) to encode coding standards for AI tools?

0No context files 1Ad-hoc or minimal 2Standardized across repos 3Comprehensive and version-controlled

2. How does your team handle progressive disclosure for large codebases when working with AI tools?

0No strategy 1Manual copy-paste of relevant files 2Structured agent docs with selective loading 3Token-budgeted context hierarchy with relevance scoring

3. Do new team members get productive with AI coding tools within their first week?

0No onboarding for AI tools 1Informal guidance from peers 2Documented setup but no workflow enablement 3Structured onboarding with context files and examples

4. Does your team use the same AI coding standards across multiple AI tools (e.g., Claude, Cursor, Codex)?

0Only one tool, no standards 1Multiple tools, each configured independently 2Shared standards adapted per tool 3Single-source rules auto-synced to all AI tools

5. Does your team encode domain expertise as reusable AI commands or prompt templates?

0No reusable commands 1Individuals save personal prompt snippets 2Team-shared command templates for common tasks 3Versioned command library with domain-specific workflows

Pillar 2: Multi-Agent Orchestration

How effectively does your team decompose work and coordinate multiple AI agents for complex tasks?

0 of 5 answered

6. Does your team use spec-driven development to guide AI code generation?

0No specs for AI 1Informal requirements in prompts 2Structured specs for major features 3Machine-readable specs (text, visual, or both) drive all AI generation

7. How does your team handle task decomposition for AI-assisted work?

0Single-prompt approach 1Manual breakdown, sequential execution 2Structured task decomposition with dependency tracking 3Automated decomposition with parallel agent execution

8. Can your team run multiple AI agents in parallel on independent tasks?

0One agent at a time 1Occasionally run 2 agents manually 2Regularly parallelize with worktrees or agent teams 3Systematic parallel orchestration with merge coordination

9. Does your team use event automation (e.g., hooks, triggers) to enforce quality when AI generates code?

0No automation around AI actions 1Manual checks after AI generates code 2Pre-commit hooks for linting and secret scanning 3Lifecycle hooks trigger validation, tests, and security checks on every AI action

10. Does your team have patterns for handling AI agent failures or stuck states?

0Start over from scratch 1Ad-hoc retry or manual intervention 2Documented recovery patterns with a clear retry limit 3Automated failure detection with approach pivot and structured handoff

Pillar 3: CI/CD Integration

How well does your pipeline handle AI-generated code with appropriate quality gates, testing, and deployment guardrails?

0 of 5 answered

11. Does your CI pipeline run automated tests on AI-generated code before merge?

0No CI pipeline 1Basic tests, often skipped for AI PRs 2Same pipeline as human code 3Enhanced pipeline with AI-specific quality gates

12. How does your team test AI-generated code?

0Manual testing only 1AI generates code, humans write tests 2TDD with AI: tests first, AI implements 3Comprehensive TDD with property-based and integration testing

13. How does your team manage security for AI-generated code and AI tool access?

0No security practices for AI 1Manual security review of AI output 2Automated scanning (SAST, dependency audit) in pipeline 3Sandboxed AI environments plus layered scanning: SAST, secrets detection, dependency audit, SBOM

14. How does your team handle code review for AI-generated PRs?

0No review process 1Same review as human code, often rubber-stamped 2Structured review with AI-specific checklist 3Multi-layer review: automated checks, human review, traceability

15. Can your team measure the impact of AI coding tools on delivery speed and code quality?

0No metrics 1Anecdotal feedback 2Basic metrics (PRs merged, time to merge) 3DORA metrics plus AI-specific tracking: velocity, quality, coverage, and anti-pattern frequency

Your score is ready. One step to unlock it.

You've assessed all 3 pillars. We'll email you a personalized scorecard with your score, per-pillar breakdown, and a 30-day action plan.

Your overall readiness score and letter grade
Per-pillar breakdown with specific patterns to apply
A sequenced 30-day priority action plan

Your results are on the way.

Sent to . Check your inbox in a minute or two. If you don't see it, check spam.

You'll occasionally hear from us about workshops. Unsubscribe anytime from any email.

Want Help Implementing Your Results?

Our workshop turns your scorecard into hands-on enablement — taught against your codebase, scoped to the gaps the assessment just surfaced.

See how a workshop addresses this →

Or book a conversation to discuss your results →