# AI use cases for Aidlc Workflows (awslabs/aidlc-workflows) Repo: [awslabs/aidlc-workflows](https://github.com/awslabs/aidlc-workflows) | Use cases: 3 | Page: https://dedupedai.com/use-cases/awslabs-aidlc-workflows/ AI-DLC is an AWS Labs set of harness-neutral workflow steering rules that turn AI coding assistants into structured, verifiable software delivery workflows across Claude Code, Kiro, Codex CLI, Cursor, opencode and GitHub Copilot. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Give your coding agent a structured delivery workflow: https://dedupedai.com/use-cases/awslabs-aidlc-workflows/add-structured-delivery-workflow-to-project/ - Use one workflow across Claude Code, Kiro, Codex and Copilot: https://dedupedai.com/use-cases/awslabs-aidlc-workflows/same-workflow-across-team-tools/ - Build and test AI-DLC harness packages from source: https://dedupedai.com/use-cases/awslabs-aidlc-workflows/build-and-test-aidlc-harness-packages/ ## How to give your coding agent a structured delivery workflow with Aidlc Workflows Repo: [awslabs/aidlc-workflows](https://github.com/awslabs/aidlc-workflows) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Install AWS Labs AI-DLC, configure it for your coding tool, and start work with /aidlc so the agent moves through phases with approval gates and an audit trail. **Who it is for:** A developer or tech lead who wants AI coding to follow a real delivery process, with requirements, design, tests and review, not one long chat. **The problem:** Ad hoc AI coding loses context as projects grow, and nobody can show how a requirement turned into the code that shipped. **The outcome:** A project with 5 phases and 33 stages, human approval gates, persistent state and a 126 event audit trail tying requirements to code and tests. ### Steps 1. Install on macOS or Linux: `curl -fsSL https://github.com/awslabs/aidlc-workflows/releases/latest/download/install.sh | sh` 2. In your project run `aidlc config --harness claude` (or cursor, codex, opencode, copilot, kiro) 3. Check the setup with `aidlc doctor` 4. Start the agent and run `/aidlc Build a REST API for inventory management`, with your own goal 5. Approve or correct each stage at its gate before it moves on 6. If plugin stages disappear after a refresh, run /aidlc plugin sync ### Prompt to paste into your agent ```text This project uses awslabs/aidlc-workflows. Run /aidlc for: . Stop at each approval gate, summarise what the stage produced and what you need from me, and keep the audit trail up to date. ``` ### Caveat The README recommends a capable reasoning model; results drop with weaker ones. Risk, remote script: The installer pipes a release script into sh; read it before running. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/awslabs-aidlc-workflows/add-structured-delivery-workflow-to-project/ ## How to use one workflow across Claude Code, Kiro, Codex and Copilot with Aidlc Workflows Repo: [awslabs/aidlc-workflows](https://github.com/awslabs/aidlc-workflows) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Configure the same AI-DLC core for each coding tool on your team, so everyone runs the same stages and gates whichever agent they prefer. **Who it is for:** An engineering team where people use different AI coding tools but the delivery process has to be the same for audits and reviews. **The problem:** Each tool gets its own prompts and rules, so the process differs by person and reviewers cannot compare work. **The outcome:** One deterministic engine behind every supported harness, with the same phases, approvals and audit trail in each tool. ### Steps 1. Install AI-DLC on each machine; on Windows use `irm https://github.com/awslabs/aidlc-workflows/releases/latest/download/install.ps1 | iex` in a normal PowerShell window 2. Each person configures their tool, for example `aidlc config --harness codex` or `aidlc config --harness cursor` 3. Kiro IDE users run `aidlc config --harness kiro-ide`; Copilot users use `aidlc config --harness copilot` 4. Everyone runs `aidlc doctor` from the project root 5. Start workflows with /aidlc, or $aidlc in Codex CLI 6. After version upgrades finish the active workflow, then run aidlc config again to avoid skew ### Prompt to paste into your agent ```text Configure awslabs/aidlc-workflows in this repo for and run aidlc doctor. Then compare the harness table in the README with what my teammates use () and list any version minimums we do not meet. ``` ### Caveat Each harness has a minimum version, for example Codex CLI 0.145.0 and opencode 1.17. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/awslabs-aidlc-workflows/same-workflow-across-team-tools/ ## How to build and test AI-DLC harness packages from source with Aidlc Workflows Repo: [awslabs/aidlc-workflows](https://github.com/awslabs/aidlc-workflows) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Use the repo's Bun scripts to regenerate harness packages from core, check that output is deterministic, and run smoke, unit and integration tests before a change. **Who it is for:** A contributor or platform engineer customising AI-DLC stages, agents or rules for their company. **The problem:** Hand editing generated files breaks on the next release, and changes to one harness can silently drift from the others. **The outcome:** Packages rebuilt from core, a determinism check that fails on drift, and test results you can attach to a pull request. ### Steps 1. Clone the repo and run `bun install --frozen-lockfile` 2. Make your change in core/ or harness/ only, never in the generated dist output, since that is overwritten 3. Generate packages with `bun scripts/package.ts` 4. Run the determinism guard: `bun scripts/package.ts --check` 5. Run tests: `bun tests/run-tests.ts --ci` 6. Before a release run `bun tests/run-tests.ts --release` ### Prompt to paste into your agent ```text In my clone of awslabs/aidlc-workflows, change in core/, regenerate packages, run the determinism check and the CI tests, and summarise any failures with the file that caused each. ``` ### Caveat Building from source needs Bun, unlike the native install. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/awslabs-aidlc-workflows/build-and-test-aidlc-harness-packages/