---
title: About Cromus — AI Workflow Intelligence Platform
description: Cromus is the first Workflow Intelligence as a Service (WIaaS) platform. We build the pre-execution intelligence layer for AI workflows — governing, costing, and optimizing before execution.
canonical: https://cromus.ai/about
source_html: https://cromus.ai/about
---

# About Cromus

> The pre-execution intelligence layer for AI workflows.

Cromus is the first **Workflow Intelligence as a Service (WIaaS)** platform. We define and own the WIaaS category — a new infrastructure layer that sits between workflow design and workflow execution.

---

## What we do

Organizations building AI workflows face the same problem: they design, execute, observe cost and quality failures, and iterate. All the tools they use — orchestration frameworks, runtime stacks, observability platforms — operate after execution begins.

Cromus operates before execution. It scores, validates, simulates, and optimizes AI workflows deterministically — with zero LLM calls — so teams understand what a workflow will cost and where it can improve before a single token is spent.

---

## Our platform

**Skill Studio** — Compile SOPs into portable SKILL.md files. Score AI-readiness. Validate against ETHOS.md behavioral policies and MEMORY.md contracts. Visualize workflow graphs. Target any AI execution framework.

**Cost Simulator** — Compare five cost-quality modes (Eco, Cost, Balanced, Quality, Open Source) across a verified registry of 62+ models from 11 providers. See per-run and monthly projections before committing.

**Croms™ Engine** — Quantify preventable waste across four dimensions (cost, latency, failure risk, structural inefficiency) and receive concrete optimization recommendations with dollar estimates.

**Open Specs** — SKILL.md, ETHOS.md, and MEMORY.md are open specifications with free public validators. Any team can adopt them regardless of whether they use Cromus.

**MCP Server** — Six deterministic workflow intelligence tools (`score_skill`, `simulate_cost`, `validate_skill`, `validate_ethos`, `validate_memory`, `reality_check_summary`) available at `https://mcp.cromus.ai/mcp` for Claude Cowork, OpenAI Codex CLI, and any MCP-compatible client.

---

## Our positioning

Cromus is the **upstream governance layer** above:
- Orchestration frameworks (LangGraph, CrewAI, AutoGen, n8n)
- Runtime agent stacks (Anthropic Claude Managed Agents, Google Agent Platform, AWS Bedrock Agents, Azure agent gateways)
- Observability tools (LangSmith, Helicone, Langfuse)

We do not compete with these tools. We produce the governed input they execute and observe.

---

## Model coverage

Verified registry of **62+ models** across **11 providers**: OpenAI, Anthropic, Google, xAI (Grok), DeepSeek, Mistral, Meta (LLaMA), Cohere, Amazon (Nova), Microsoft (Phi), and Qwen — with weekly pricing and deprecation checks.

---

## Contact

Questions, partnerships, press, or enterprise inquiries: [Contact us →](/contact)

Admin: aecardonac@gmail.com

---

## Related pages

- [Workflow Intelligence as a Service →](/workflow-intelligence-as-a-service)
- [Pre-Execution Governance →](/pre-execution-governance)
- [Pricing →](/pricing)
- [Manifesto →](/manifesto)