Our take
Activeregular commits and releases- Activity
- 85
- Community
- 61
- Issues
- 70
- Pull requests
- 84
- Merges more pull requests than 96% of the projects we track
- Releases more often than 86% of the projects we track
AI-DLC is an actively maintained, cross-harness workflow system with a substantial contributor base, suited to teams that want more control and traceability in AI-assisted delivery.
Good fit if
- You use one of its supported harnesses and want AI work organized around explicit stages, decisions, and approvals.
- Your team values persistent workflow context and audit evidence across feature, bug-fix, infrastructure, or security work.
- You want a shared methodology across coding-agent tools, while leaving model and provider choices to each harness.
Look elsewhere if
- You want a standalone coding agent or a model-provider-specific development environment. AI-DLC supplies workflow steering, not a replacement for the harness.
- Your process is intentionally unstructured and does not need approval gates, persistent workflow records, or defined stages.
- You need a harness that is not among the supported integrations and do not intend to port one.
All health signals
| Last commit | 2026-10-20 (today) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 45 (latest v2.11.1, 2026-10-20) |
| Contributors | 74 (top contributor: 71% of commits) |
| Issues closed, last 90 days | 6+ (typically closed in 1 day) |
| Pull requests merged, last 90 days | 95 (typically merged in 0 days) |
| Project age | 11 months |
Checked on 2026-10-20 with the GitHub API.
Overview
AI-DLC (AI-Driven Development Life Cycle) is a harness-neutral core for turning AI coding assistants into structured, verifiable software-delivery workflows. It addresses the loss of context and process in ad-hoc AI coding by connecting requirements, decisions, implementation, tests, and operational work across a lifecycle.
The same deterministic engine works with Claude Code, Kiro CLI and IDE, Codex CLI, Cursor, opencode, and GitHub Copilot. Teams can use it to guide work through approval gates and keep an audit trail, while retaining the model and provider setup of their chosen harness.
Key Features
- Organizes work into five phases and 33 stages, from initialization through operation.
- Provides 11 workflow profiles for work such as features, bug fixes, infrastructure, security, and proofs of concept.
- Uses 14 agents, including domain experts, reviewers, and an adaptive composer.
- Pauses at human approval gates and records source-bound review evidence.
- Maintains persistent workflow state, team knowledge, learned rules, and an audit trail.
- Offers a native
aidlccommand for installation, project configuration, diagnostics, and lifecycle management. - Supports multiple coding-agent harnesses through one shared methodology and harness-specific integrations.
Use Cases
- A product team can use a feature workflow to have an AI assistant clarify requirements and obtain approval before implementation proceeds.
- A developer can use a focused bug-fix or security workflow when an unstructured coding prompt is not enough to track decisions and review evidence.
- A platform or infrastructure team can guide AI-assisted infrastructure work through defined stages and approval points.
- A team supporting several coding-agent tools can share one workflow methodology across those harnesses rather than maintaining separate processes.
What you need
Detected in the repository
- Node.js (from package.json)
- Python >=3.12 (from pyproject.toml)
- A test suite and automated checks on GitHub Actions
License in plain words
MIT-0
Read the LICENSE file before using it in a product.
Getting Started
On macOS, Linux, or WSL, install the latest stable release:
curl -fsSL https://github.com/awslabs/aidlc-workflows/releases/latest/download/install.sh | sh
Then configure a project and start a workflow, for example:
aidlc config --harness claude
aidlc doctor
Open the project in the configured harness and invoke /aidlc with a description of the work. Codex CLI uses $aidlc. See the README and Getting Started guide for Windows installation, other harnesses, and setup details.
Alternatives
- compound-engineering-plugin: Compound Engineering packages reusable agent skills for planning, implementation, review, and learnings, while AI-DLC defines gated delivery workflows across agent hosts.
- ralph-orchestrator: Ralph Orchestrator runs agents in structured execution loops with specialized roles and quality gates, rather than providing a harness-neutral delivery workflow system.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| aidlc-workflows | TypeScript | MIT-0 | 5.4k | Active |
| compound-engineering-plugin | TypeScript | MIT | 25.4k | Active |
| ralph-orchestrator | Rust | MIT | 3.2k | Active |
Considerations
- AI-DLC steers coding assistants; generated output and costs still require human review, as the project cautions.
- The native installer avoids requiring Bun or Node.js for normal use. The repository's development requirements list Node.js and Python 3.12 or later, and its manual-copy installation path requires Bun.
- This is a fast-moving project: it has recent commits and releases, with at least 100 commits in the last 90 days and many contributors. Its contributor activity is not concentrated in one person, but teams should still evaluate releases and workflow changes as they adopt updates.
- The repository reports many open issues, so teams should check issue discussions for problems relevant to their harness or workflow before relying on a particular behavior.
- The project is licensed under MIT-0. Review the license and your organization's software policies before adopting it.
Found this useful?
Share it with someone who would like aidlc-workflows.