Skip to content

AI codes fast.
AIDLC makes it ship right.

A real software lifecycle for AI coding agents: phases, gates, state, and traceability — from idea to maintenance, on any agent platform.

npx @rasensio/aidlc init

All 7 phases with adversarial review gates

  1. ideation
  2. requirements
  3. design
  4. implementation
  5. testing
  6. deployment
  7. maintenance
aidlc — zsh
$ aidlc start full-feature --name dark-mode
✓ instance created: dark-mode
✓ branch: feat/20260821-dark-mode
→ phase: ideation

.aidlc/state/…/instance.yaml

name: dark-mode
template: full-feature
current_phase: ideation
scope: full
1/7

The problem isn’t speed

Agents produce code faster than any team can review it. What they don’t produce: requirements anyone agreed to, designs anyone challenged, tests that map to acceptance criteria, or a record of why anything was decided. AIDLC adds the missing lifecycle — without slowing the agent down.

How it works

Phases with gates

Seven phases from ideation to maintenance. Each one exits through a gate: required artifacts, adversarial review, test-per-criterion. Agents cannot skip ahead.

State on disk

The lifecycle lives in YAML under .aidlc/ — readable by any agent, any session, any worktree. Claims prevent two sessions trampling each other.

Traceability built in

Every acceptance criterion has an ID. Every task references its criteria. Every test references both. The gate fails on any criterion without a test.

It closes the loop

aidlc metrics
DORA-style roll-ups from your own state: change failure rate per template, lead time per phase, true feature cost including the bugfixes it caused.
aidlc cost
Per-instance cost tracking with metered capture — know what a feature actually cost across every session that touched it.
caused_by
Bugfixes link back to the feature that caused them. Quality leaks become measurable, not anecdotal.

Latest release

the whole trail →

v1.28.0 ·

The cheapest token is the one you never carry

Agentic coding bills you for memory, not for thought. 95.9% of the tokens this project has ever spent were re-reads of a conversation it was already carrying. Here is where the money goes, what AIDLC does about it, and the two places it does not work yet.

One command. Any agent.

Works with Claude Code, Kiro, and any platform that reads skills. The lifecycle is plain YAML in your repo — no server, no lock-in.

npx @rasensio/aidlc init
or start with a tutorial →