{"slug": "aidlc-adaptive-workflow-steering-rules-for-ai-coding-agents", "title": "aidlc: Adaptive workflow steering rules for AI coding agents", "summary": "AWS Labs released AI-DLC, a harness-neutral workflow tool that runs natively in Claude Code, Kiro CLI, Kiro IDE, Codex CLI, Cursor, opencode, and GitHub Copilot, installable via a curl or PowerShell script from the awslabs/aidlc-workflows GitHub releases. The native installer sets up the `aidlc` command and all harness runtimes without requiring Bun or Node.js, while a manual-copy path using `aidlc-copy-runtime-X.Y.Z.tar.gz` is the only route that needs Bun. Developers configure a harness with `aidlc config --harness <name>` and invoke work with `/aidlc` (or `$aidlc` in Codex CLI), with AI-DLC selecting a workflow and stopping at approval gates.", "body_md": "AI-DLC (AI-Driven Development Life Cycle) turns AI coding assistants into structured, verifiable software-delivery workflows. One harness-neutral core runs natively in Claude Code, Kiro CLI, Kiro IDE, Codex CLI, Cursor, opencode, and GitHub Copilot.\n\nThe Quick Start below installs the latest stable AI-DLC release.\n\nmacOS, Linux, or WSL:\n\n```\ncurl -fsSL https://github.com/awslabs/aidlc-workflows/releases/latest/download/install.sh | sh\n```\n\nWindows PowerShell:\n\n```\nirm https://github.com/awslabs/aidlc-workflows/releases/latest/download/install.ps1 | iex\n```\n\nThe installer sets up everything AI-DLC needs: the native `aidlc` command and\nevery harness runtime. You do not need to install Bun or Node.js; a native\ninstall runs its hooks and commands through `aidlc` and does not call `bun`.\nOn Windows, it installs for the current account\nand automatically registers the bin directory in User PATH. Run it from a normal\nPowerShell window; one opened with \"Run as administrator\" gets a warning and a\nprompt, since installing as administrator is less safe. Open a new terminal\nif another session cannot find `aidlc`. To skip both persistent and\ncurrent-process PATH changes, use\n[`-NoModifyPath`](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/18-install-and-lifecycle.md#windows-powershell).\nWindows uninstall removes only the User PATH entry recorded as installer-owned.\nOn macOS, Linux, or WSL, follow the installer's PATH instruction if needed.\n\nPrefer to manage the project files yourself, or cannot install a native\nexecutable? The manual-copy path is the one place Bun is needed: install\n[Bun](https://bun.sh/), download `aidlc-copy-runtime-X.Y.Z.tar.gz` from the\n[release](https://github.com/awslabs/aidlc-workflows/releases/latest), and copy\nthe complete `runtime/<harness>/` directory into your project. This path does\nnot require the native `aidlc` command.\n\nFrom the project root, select the harness you use:\n\n```\ncd /path/to/your-project\naidlc config --harness claude\naidlc doctor\n```\n\nReplace `claude` with `kiro`, `kiro-ide`, `codex`, `cursor`, `opencode`, or\n`copilot`. Running `aidlc config` without `--harness` starts the interactive\nsetup when a terminal is available. If you use Kiro IDE's own terminal in a\nproject folder you have not trusted yet, Kiro first asks whether you trust it.\nChoose **Trust Folder & Continue** only for your own project or one you have\nchecked, because trusting lets the folder's `.kiro` hooks run commands on your\nmachine; otherwise choose **Cancel** and review the folder first (see\n[First run](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/kiro-ide.md#first-run)).\n\nOpen your harness in the configured project and describe the work:\n\n```\n/aidlc Build a REST API for inventory management\n```\n\nCodex CLI uses `$aidlc` instead of `/aidlc`. In Kiro IDE, first choose **aidlc**\nin the chat panel's agent picker. AI-DLC selects a workflow from the request,\nasks for missing decisions, and stops at approval gates before moving forward.\n\nFor provider setup, trust prompts, and harness-specific prerequisites, use the\nguide in the table below. The complete walkthrough is in\n[Getting Started](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/01-getting-started.md).\n\n| Harness | Configure | Open | Invoke | Guide | \n|---|---|---|---|---|\n| Claude Code | `aidlc config --harness claude` | `claude` | `/aidlc` | [Getting Started](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/01-getting-started.md) | \n| Kiro IDE >= 1.1.70 / Kiro CLI >= 2.24.1 (v3) | `aidlc config --harness kiro-ide` | Open the project in Kiro IDE and choose **aidlc** in the chat panel's agent picker, or run`kiro-cli` | `/aidlc` | [Kiro IDE](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/kiro-ide.md) | \n| Kiro CLI >= 2.6 (legacy) | `aidlc config --harness kiro` | `kiro-cli chat` | `/aidlc` | [Kiro CLI](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/kiro-cli.md) | \n| Codex CLI >= 0.145.0 | `aidlc config --harness codex` | `codex` | `$aidlc` | [Codex CLI](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/codex-cli.md) | \n| Cursor | `aidlc config --harness cursor` | Open Cursor or run `agent` | `/aidlc` | [Cursor](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/cursor.md) | \n| opencode >= 1.17 | `aidlc config --harness opencode` | `opencode` | `/aidlc` | [opencode](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/opencode.md) | \n| GitHub Copilot CLI >= 1.0.74 / VS Code >= 1.130 | `aidlc config --harness copilot` | Copilot CLI or VS Code | `/aidlc` | [GitHub Copilot](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/copilot.md) | \n\nModel-provider setup belongs to the harness. Shipped project configuration\nkeeps the provider and model already selected by the user. `aidlc config providers` can apply Amazon Bedrock settings on supported project surfaces or\nrecord manual setup for other harnesses. Kiro CLI and Kiro IDE need no provider\nanswer because model access comes with Kiro. On Kiro CLI 2.24.1 or later, use\n`kiro-ide`, which pins Kiro's v3 engine; `kiro` is the legacy row ([Kiro CLI](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/kiro-cli.md)\nsays how an existing `kiro` project switches). The methodology itself is\nprovider-independent.\n\nAI-DLC works best with capable reasoning models. The current recommended model\nis Claude Opus 4.8. If Opus is not available to you, or you are not sure where\nto set the model and effort on your harness, see\n[Choosing a Model and Effort](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/18-install-and-lifecycle.md#choosing-a-model-and-effort).\n\nAd-hoc AI coding loses context as projects grow. AI-DLC keeps requirements, decisions, implementation, tests, and operational work connected through one audited lifecycle:\n\n- 5 phases and 33 stages from initialization through operation\n- 14 agents: 11 domain experts, 2 reviewers, and an adaptive composer\n- 11 workflow profiles for features, bug fixes, infrastructure, security, proofs of concept, enterprise delivery, and other common work\n- Human approval gates and source-bound review evidence\n- 119-event audit trail plus persistent state, team knowledge, and learned rules\n- The same deterministic engine across every supported harness\n\nStart with [Workflow Profiles](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/workflow-profiles.md) to compare\nClassic, Express, and the focused workflows. See the\n[AI-DLC Workflows 2.0 Specification](https://github.com/awslabs/aidlc-workflows/blob/main/assets/AI-DLC-Workflows-2.0-Specification.pdf)\nfor the architecture and methodology.\n\nImportant\n\nGenerative AI can make mistakes. Review generated output and costs before\nacting on them. See the [AWS Responsible AI Policy](https://aws.amazon.com/ai/responsible-ai/policy/).\n\n| Guide | Use it when | \n|---|---|\n| [Getting Started](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/01-getting-started.md) | Installing, configuring, and running your first workflow | \n| [User Guide](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/00-introduction.md) | Using workflows, profiles, agents, knowledge, and approval gates | \n| [Harness guides](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/README.md) | Handling provider, trust, and runtime differences | \n| [Install and Lifecycle](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/18-install-and-lifecycle.md) | Updating, pinning, installing offline, using mirrors, or uninstalling | \n| [Harness Engineer Guide](https://github.com/awslabs/aidlc-workflows/blob/main/docs/harness-engineering/00-overview.md) | Reshaping stages, agents, rules, sensors, and knowledge | \n| [Development and Releases](https://github.com/awslabs/aidlc-workflows/blob/main/DEVELOPERS.md) | Taking a PR through AI review, preview testing, and stable publication | \n| [Developer Reference](https://github.com/awslabs/aidlc-workflows/blob/main/docs/reference/00-overview.md) | Changing the engine, hooks, packaging, or tests | \n\n- `core/` - hand-authored, harness-neutral methodology and engine\n- `core/tools/` - 93 aidlc-*.ts engine and authoring tools\n- `harness/<name>/` - thin, harness-specific manifests and integrations\n- `plugins/<name>/` - optional AIDLC plugins\n- `scripts/` - packaging, binary, installer, and release tooling\n- `tests/` - smoke, unit, integration, and end-to-end tests\n- `docs/` - user, harness-engineering, and developer documentation\n- `dist/` and`dist-release/` - generated, ignored local outputs\n\nEdit `core/` or `harness/<name>/`, never generated `dist*` output.\n\nInstall dependencies and generate every harness:\n\n```\nbun install --frozen-lockfile\nbun scripts/package.ts\n```\n\nUseful commands:\n\n```\nbun scripts/package.ts <name>     # generate one harness\nbun scripts/package.ts --check    # determinism guard\nbun tests/run-tests.ts --ci       # smoke, unit, and integration\nbun tests/run-tests.ts --release  # full release acceptance\n```\n\nSee the [Contributing Guide](https://github.com/awslabs/aidlc-workflows/blob/main/docs/reference/11-contributing.md) for the complete\ndevelopment workflow and [Porting to a New Harness](https://github.com/awslabs/aidlc-workflows/blob/main/docs/harness-engineering/09-porting-to-a-new-harness.md)\nto add another runtime.\n\nRun `aidlc doctor` from the project root first. Common fixes:\n\n| Symptom | Fix | \n|---|---|\n| `aidlc` is not found | On Windows, quit and reopen the terminal app or IDE (a new tab inside one that was open during the install keeps the old PATH), or use the direct command printed with [`-NoModifyPath`](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/18-install-and-lifecycle.md#windows-powershell) . On Unix, apply the installer's PATH instruction. | \n| Project/runtime version skew | Finish the active workflow, then run `aidlc config` | \n| Codex hooks do not run | Trust the project hooks as described in the [Codex guide](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/harnesses/codex-cli.md) | \n| Bedrock access fails | Enable the configured models and verify AWS credentials and region | \n| Plugin stages disappear after refresh | Run `/aidlc plugin sync` | \n| Refreshed skills do not take effect | Start a new harness session | \n\nSee [Troubleshooting](https://github.com/awslabs/aidlc-workflows/blob/main/docs/guide/15-troubleshooting.md) for diagnostic and\nrecovery procedures.", "url": "https://wpnews.pro/news/aidlc-adaptive-workflow-steering-rules-for-ai-coding-agents", "canonical_source": "https://github.com/awslabs/aidlc-workflows", "published_at": "2026-10-15 07:20:39+00:00", "updated_at": "2026-10-15 07:47:45.427623+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "ai-tools", "ai-products"], "entities": ["AWS Labs", "AI-DLC", "Claude Code", "Kiro CLI", "Kiro IDE", "Codex CLI", "Cursor", "GitHub Copilot"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/aidlc-adaptive-workflow-steering-rules-for-ai-coding-agents", "markdown": "https://wpnews.pro/news/aidlc-adaptive-workflow-steering-rules-for-ai-coding-agents.md", "text": "https://wpnews.pro/news/aidlc-adaptive-workflow-steering-rules-for-ai-coding-agents.txt", "jsonld": "https://wpnews.pro/news/aidlc-adaptive-workflow-steering-rules-for-ai-coding-agents.jsonld"}}