What changes
Before and after
The classic way
- Requirements live in tickets, chats and slides; the code is the only complete description of what was built.
- Estimates are made by hand and the backlog is sliced by people's availability.
- Which AI tools may be used is unsaid, so every team decides alone.
The AI way
- Intent is written as a spec that agents read as their brief and people read as the contract. GitHub's Spec Kit calls the phases Spec, Plan, Tasks and Implement, each producing a Markdown artifact that feeds the next [6].
- The organization states its AI stance in writing: expectations for AI use, support for experimentation and which tools are permitted. DORA found this to be the first of seven capabilities that magnify AI's positive impact [2].
- Work is cut into small batches on purpose, because small batches amplify AI's positive influence on product performance and reduce friction [2].
Step by step
The three steps of Plan
- 1
Intent arrives Person
A product owner writes what should change and why, in a spec, not a chat message. The spec is the first artifact of the lifecycle and the one everything later is checked against.
- 2
The agent drafts the plan AI
An agent reads the spec and the repository, proposes a technical plan and a task list, and flags what the spec leaves open. Spec Kit's phases name exactly this hand-off: Spec, then Plan, then Tasks [6].
- 3
A person owns the spec Gate
Nothing moves until a named person accepts the spec and the plan. The agent can propose; it cannot decide what the product is.
The gate
A named owner accepts the spec and the plan. The spec is version-controlled next to the code, so every later change can be traced to the sentence that asked for it.
What the AI does here
Drafts plans and task lists from the spec, finds gaps and contradictions, keeps the spec and the code in step as both change.
Tools people use at this stage
- GitHub Spec Kit
- Claude Code (CLAUDE.md)
- AGENTS.md
- Kiro
- Copilot
Named as examples, not endorsements; the center is not affiliated with any of them.
What this stage adds
The risks, and the practices that answer them
Each risk is a weakness in the Learning Center, with the incidents where it happened; each practice has its sources there.
Risks
- Overreliance on AI outputA plan that reads well is accepted without being checked.
- Poisoned agent configuration and instruction filesThe files that brief the agent become a way to steer it.
Read for this stage
Sources
- [6] GitHub Spec Kit: spec-driven developmentread 2026-10-27
- [2] Google Cloud: Introducing the DORA AI Capabilities Modelread 2026-10-27
- [1] Google Cloud: Announcing the 2025 DORA State of AI-assisted Software Development reportread 2026-10-27