AIWG
Carry decisions forward and review complex work in the AI tools you already use. AIWG supplies reusable instructions and workflows that help your assistant produce inspectable results and save context for the next task.
Open your project in your AI tool and follow Install, Connect, and Verify. The agent should preserve your existing work, verify the connection, and report whether your tool needs a reload.
Once connected, ask for a README review:
Use AIWG to review this project's README for unclear positioning and missing
onboarding steps. Save a report at
.aiwg/marketing/brand/audit/readme-review.md with file references and the
three highest-priority fixes. Leave the README unchanged.
Open the report. Check that its findings identify real passages, explain the reader's problem, and recommend specific changes. In a later session, ask the agent to read the report and implement the first agreed fix.
Follow the quickstart or see an illustrative result and alternative tasks.
Plan a change, review its implementation, and connect decisions to tests.
Prepare a campaign brief, review messaging, and refine a draft.
Organize sources, trace citations, and keep findings available for later work.
Explore security reviews, forensics, media curation, operational runbooks, and optional utilities.
AIWG places readable instructions in locations your provider supports. Workflows can save plans, findings, and decisions in the project's .aiwg/ directory. Your assistant still needs to consult those artifacts, and its conclusions still need checking.
Integration status, deployment scope, and native capabilities vary. Consult the provider inventory and provider setup guides for your tool. Optional services and external actions require their corresponding configuration and integrations.
AIWG has 19 named provider integrations:
| Platform | Status |
|---|---|
| Claude Code | Stable |
| OpenAI Codex | Stable |
| GitHub Copilot | Stable |
| Cursor | Stable |
| DeepSeek Harness | Experimental |
| Factory AI | Stable |
| Flue | Experimental |
| Grok Bot | Stable |
| Grok Build | Experimental |
| Muse Code | Stable |
| Hermes | Stable |
| OpenCode | Stable |
| OpenClaw | Stable |
| OpenHuman | Experimental |
| Google Antigravity CLI | Experimental |
| Oh My Pi | Experimental |
| Pi Coding Agent (pi.dev) | Experimental |
| Warp Terminal | Stable |
| Devin Desktop | Stable compatibility adapter |
Examples, building blocks, and practical limitations.
Assess adoption effort, review quality, and cost on a bounded task.
See deployment, project artifacts, and optional runtime components.
Exact commands for manual operation and automation.