Lifecycle

Overview

The lifecycle model describes six phases that engineering work moves through, from specification to deployment. Each phase has handoff triggers, constraints, and checklists that guide quality transitions.

Phases are conceptual workflow vocabulary. They describe what kind of work happens and which discipline applies. No YAML file backs them as a data entity.

The Six Phases

flowchart LR
    specify["Specify"] --> plan["Plan"]
    plan --> scaffold["Scaffold"]
    scaffold --> code["Code"]
    code --> review["Review"]
    review --> deploy["Deploy"]

Phase	Purpose
specify	Define WHAT users need and WHY
plan	Design HOW to build (architecture, technical choices)
scaffold	Prepare dev environment, dependencies, credentials
code	Implement the solution and write tests
review	Verify implementation against acceptance criteria
deploy	Ship to production, monitor CI/CD

Handoffs

Handoffs are the transitions between phases. Each phase defines triggers that specify where work flows next.

Specify Phase

Handoff	Target	Trigger
Refine/Alternative Spec	specify	Requirements need rework
Plan	plan	Spec accepted

Plan Phase

Handoff	Target	Trigger
Refine/Alternative Plan	plan	Design needs rework
Scaffold	scaffold	Plan accepted

Scaffold Phase

Handoff	Target	Trigger
Retry Setup	scaffold	Environment setup failed
Update Plan	plan	Plan needs revision based on setup findings
Code	code	Environment ready

Code Phase

Handoff	Target	Trigger
Request Review	review	Implementation complete, tests pass

Review Phase

Handoff	Target	Trigger
Request Changes	code	Issues found, changes needed
Needs Re-planning	plan	Fundamental problems require redesign
Deploy	deploy	Review approved

Deploy Phase

Handoff	Target	Trigger
Fix Pipeline	deploy	CI/CD issues need resolution

Constraints

Each phase defines constraints that limit what actions are allowed. Constraints are especially important for AI agents. They prevent scope creep. They make sure agents stay within their authorized boundaries.

Phase	Cannot
specify	Write code, make architectural decisions
plan	Commit code, go beyond specified requirements
scaffold	Implement features, change the plan
code	Change architecture, deviate from the plan
review	Add features, only verify and request changes
deploy	Change code, only fix pipeline configuration

Checklists

Checklists make sure quality holds at phase transitions. Each skill defines its own checklist items as flat fields in the agent section of the skill's YAML definition.

Two Types

Type	When	Purpose
Read-Then-Do	Before you start work	Prerequisites and preparation
Do-Then-Confirm	Before you hand off	Verification criteria

Read-then-do checklists are entry gates. Read each item, then do it. Do-then-confirm checklists are exit gates. Do them from memory, then confirm every item before you cross a boundary.

How Checklists Are Defined

Skills define checklists as flat fields in their agent section:

agent.readChecklist — items for the read-then-do gate
agent.confirmChecklist — items for the do-then-confirm gate

Both fields are simple arrays of strings on the skill's agent section. They are not keyed by proficiency level. They do not come from a phase x skill matrix. Every skill that ships an agent: section must define both checklists.

Example

Given a practitioner-level CI/CD skill:

Read-Then-Do (before you code):

Review the deployment pipeline configuration
Understand the test infrastructure
Check branch protection rules

Do-Then-Confirm (before you request review):

All new code has test coverage
Pipeline passes on the feature branch
Documentation is updated for changed interfaces

Phases and Agents

Agents are generated per discipline and track. Each combination gets one agent. A lifecycle phase does not get its own agent. Phases guide an agent's workflow focus: what to prioritize, what constraints apply, and which checklist items are relevant at each point in the work.

Generate an agent profile for a discipline and track:

npx fit-pathway agent <discipline> --track=<track> --output=./agents

The generated profile includes the agent's full skill set, checklist items, and behavioural expectations. Phases are not an input to agent generation. They describe the workflow context in which the agent operates.

Run npx fit-pathway agent --help for the full command surface.

What's next

Core Model

How disciplines, levels, tracks, skills, and behaviours combine to produce complete role definitions.

YAML Schema Reference

File format reference for every entity type, with examples and links to published JSON Schema and RDF/SHACL definitions.