Platform · Governed AI
AI Agents with Business Guardrails
AI acting on a large operation can't hallucinate a price or break an approval limit. CWS Platform's 6 AI Agents consume the deterministic Core via MCP, under exactly the same rules as a human seller, within the Autonomy Gradient. They suggest and execute the reversible; the human decides the rest.
Autonomy is the company's configuration, not the agent's promise
The matrix with the 6 AI Agents as rows and the three bands as columns, stating action by action what each one runs alone, what it proposes and what it never does.
- Action by action, not agent by agentThe same agent shows up across all three bands, because what sets the band is the consequence of the action, not how much it is trusted.
- The matrix belongs to the companyChanging a band is a logged configuration review, and none of the six agents can edit this screen.
- The Core outlives the AIAll six consume the deterministic Core via MCP under the same rules as a human seller, and if the AI goes down the Core keeps operating.
This is the AI Agents page showing its own mechanism instead of describing it. Each action's band is a company decision, changing it is a logged configuration review, and no agent edits this matrix. The counter of rules altered by an agent reads zero in every band, which is the verifiable claim behind governance by infrastructure rather than by prompt.
Rules Auditor Agent
Audits 1,029 parameters in 45s, detects conflicts, prevents leakage. It suggests, the human decides.
View agent →Sales Copilot
Account ranking, margin and inactivity alerts, builds the quote.
View agent →Procurement Agent
Semantic search, multi-seller comparison, builds the cart, reorders.
View agent →Merchandising Agent
Monitors the store, incomplete listings, suggests campaigns.
View agent →Catalog Enrichment Agent
A 7-step pipeline, from photo to publication per SKU.
View agent →Replenishment Agent
Redistributes across DCs, stockout alerts, suggests purchases.
View agent →Autonomy Gradient
AI governance by infrastructure, not by prompt.
- Green: the agent executes on its own (reversible, low-risk action).
- Yellow: the agent proposes, the human confirms with one click.
- Red: the agent never acts. It informs and escalates to the human.
Frequently asked questions
Can the AI Agents change a business rule?
No. The AI Agents consume the deterministic Core via MCP (Model Context Protocol), under exactly the same rules as a human seller. They interpret intent and suggest; the rule is executed by the Commerce Rules Engine, and the human decides what escalates.
What is the Autonomy Gradient?
It's AI governance by infrastructure, not by prompt. Each agent action falls into one of three zones: green (executes on its own, low risk), yellow (proposes, the human confirms) or red (never acts, escalates to the human).
If the AI goes down, does the operation stop?
No. The deterministic Core executes the rules by code, at zero inference cost. The AI Agents orchestrate that Core, but if the AI goes down, negotiation keeps running on the Core.