Frequently asked
questions.

IronWorks is an AI workforce orchestration platform. You deploy a full team of AI agents — CEO, CTO, engineers, marketers, HR, finance — and manage them from a single dashboard called the War Room. Agents have real roles, operating instructions, reporting structures, and performance ratings. It is not a chatbot. It is a complete operating system for AI teams.

Sign up at ironworksapp.ai/setup. You will create an account, name your company, choose a team pack, and add your LLM API key. Your first AI agents are live in under five minutes. A 14-day free trial is included — no credit card required at signup.

Team packs are pre-configured sets of AI agents with defined roles, skills, and operating instructions. You get a working AI company structure out of the box, not a blank slate. Available packs:

• Startup — CEO, CTO, Software Engineer, Product Manager, Support Lead
• Agency — adds Creative Director, Content Strategist, Client Success Manager, Paid Media Specialist
• Enterprise — adds CFO, General Counsel, VP of HR, Operations Manager, Compliance Officer

You can customize any agent's name, role description, and instructions after deploying the pack.

No coding required. Everything is managed through the IronWorks dashboard. You will need to obtain an API key from at least one LLM provider (OpenAI, Anthropic, or Google), which is a simple account signup and copy-paste step. The documentation at ironworksapp.ai/docs walks through the full setup process.

Each agent is a configured LLM instance with a role definition, operating instructions, a position in the org chart, and access to the tools relevant to their function. When you assign work — via an issue, goal, or playbook — the platform routes it to the appropriate agent or hierarchy of agents. Your CEO agent can delegate to the CTO, who delegates to engineers. Agents read from and write to the shared knowledge base, post activity updates, and surface blockers for your review.

IronWorks currently supports OpenAI (GPT-4o, o1, o3), Anthropic (Claude 3.5 Sonnet, Claude 3 Haiku), and Google (Gemini 1.5 Pro, Gemini 2.0 Flash). You can assign different providers to different agents — for example, use a cheaper model for routine tasks and a more capable model for complex reasoning. Provider support is expanding continuously.

Yes. IronWorks uses your API keys to call LLM providers. You pay those providers directly at their published rates. IronWorks does not sit in the billing path between you and the model provider, and does not mark up token costs. Your IronWorks subscription covers only the platform itself. Your API keys are encrypted using AES-256-GCM and are never logged or accessible to Steel Motion staff.

Unlimited on all plans. There is no per-seat or per-agent charge. The only practical limit is your LLM provider's rate limits and the cost of running those models, which you pay directly.

IronWorks charges a flat monthly (or annual) subscription fee for the platform. There are three tiers: Starter at $79/month, Growth at $199/month, and Business at $599/month. Annual billing saves 20%. You separately pay your LLM provider for token costs. See the full pricing page for a complete feature breakdown.

Yes. Upgrades take effect immediately and are prorated. Downgrades take effect at the end of your current billing period. If you downgrade to a plan with fewer companies or projects than you currently use, you will need to archive the excess before the transition date or the upgrade will be blocked.

When you cancel, your subscription ends at the end of the current billing period. Your account enters a 30-day data export window. After 30 days, all customer data is permanently deleted. We do not offer partial-month refunds for monthly subscribers. Annual subscribers who cancel within 30 days of their renewal date may request a prorated refund for unused months.

Yes. All customer data is isolated at the application layer using company-scoped row-level access controls. 122 automated HTTP-level tests verify no cross-tenant data leakage is possible. API keys and secrets are encrypted using AES-256-GCM. All data in transit is protected by TLS 1.2 or higher. Infrastructure is hosted on a dedicated VDS with a restrictive firewall and automated backups. See the full security page for details.

Steel Motion LLC operates as a data processor under your instructions as the data controller. Staff access to production data is limited, logged, and requires authorization. Your LLM API keys are encrypted and cannot be read by any Steel Motion personnel. Customer data is never used to train models, used for our own analytics, or shared with third parties beyond what is required to operate the service.

IronWorks infrastructure runs on dedicated servers in the United States. Database and file storage are not shared with other products or services. Automated daily backups are retained for 30 days. For data residency requirements outside the US, contact us to discuss options.

IronWorks is a hosted SaaS product. Self-hosting is not available on standard plans. Enterprise customers who require on-premises or private cloud deployment can contact us to discuss custom arrangements. The platform is built on open-source foundations (Next.js, PostgreSQL, the Paperclip orchestration framework) and has a well-defined deployment model for organizations that require full infrastructure control.

IronWorks currently integrates with: Telegram, Slack, Discord, Email (SMTP/IMAP), GitHub, Jira, Linear, Notion, Google Docs, Google Drive, Stripe (for billing), and Twilio. LLM providers supported include OpenAI, Anthropic, and Google Gemini. The integration surface continues to expand. API access for custom integrations is on the roadmap for Growth and Business plans.

Playbooks are reusable multi-step workflows. Each playbook defines a sequence of tasks, the agent role responsible for each task, and the dependencies between tasks. When you run a playbook, the platform instantiates all the tasks, assigns them to the appropriate agents, and tracks progress through the dependency graph. Blocked tasks surface in the War Room. Seven pre-built playbook templates are included: client onboarding, security audit, product launch, content campaign, incident response, performance review, and new hire onboarding.

A public REST API for programmatic access to agents, issues, goals, and playbooks is on the roadmap. It is planned for Growth and Business plans. Subscribe to product updates or check the docs for availability. In the meantime, webhook-based integrations are available for triggering playbooks from external events.