Patterns

How agents work with ShipMore — the conventions that hold whether a human is driving Claude Code interactively or an autonomous agent is running on a schedule.

Drafts by default

Agents create content as drafts unless explicitly told to publish. Humans review and publish from the admin. Every agent action is versioned, auditable, and reversible through version history — so an agent can move fast without putting the live site at risk.

Tenant-scoped

Every agent action targets one tenant explicitly. An agent working on one tenant can’t reach across into another tenant’s data — isolation is enforced by the platform, not left to the agent to get right.

One model, every interface

Whether an instruction arrives over MCP, the CLI, the admin, or a webhook, it drives the same underlying operations and behaves identically. An agent doesn’t have to learn a different mental model per interface.

Two operating modes

Human-on-demand (Claude Code / Cursor)

"Build me a landing page for an AI image generator that sells credit packs"
"Add a pricing section using our existing Stripe products"
"Create a new tenant for client Acme with their domain and branding"

Autonomous agent (scheduled / event-driven)

Every Monday    → generate + publish weekly blog post from trending topics
On lead signup  → update lead status, trigger welcome email sequence
On Stripe event → price changed? update pricing block copy automatically
Nightly         → check for expired promos, flag or auto-fix

Both modes drive the same operations; the scheduling and trigger infrastructure is just how the instruction gets in.