Persona · Agencies

KERNSTACK for Agencies

Agencies choose KERNSTACK because delivery repeats: every client gets the same declared shape — project, preview targets, client conversation, per-client meter — instead of a snowflake assembled on whichever vendor the last project used. Clients review running previews, approve in the channel they actually read, and at handoff they receive the project itself: repository, declaration, data, and history.

Hosted Git Agent Sandboxes Collaboration Board stack.json IaC

Talk to us Read Technical Documentation → Software Lifecycle Cloud →

Why Agencies outgrow assembled cloud and tool collages

Studios delivering client projects repeatedly — without adopting each client’s cloud account, or leaving credentials in a password manager as the handoff.

A delivery shape, not a fresh invention per client

The studio’s template stack.json carries the accumulated lessons: how routes are named, how content publishes, where uploads land, which approvals gate a launch. A new client starts from the template on European infrastructure the studio can name in the pitch, and the differences between clients become the visible part — their product — instead of invisible plumbing folklore.

Clients inside the process, at arm’s length from the machinery

Bound project conversations (mirrored to the client’s Telegram or WhatsApp) give feedback a home with states; agent-made changes reach the client as demo links, never as surprises on their site. Per-client projects mean per-client meters, so the invoice conversation uses facts. The studio looks organised because, structurally, it is.

Per-client credit balances: true delivery cost without allocation spreadsheets
Approvals happen against live previews, recorded in the project — scope disputes lose their favourite evidence gap
Handoff transfers a working, declared project — not a zip file and a credentials doc

stack.json — Agencies Stack plan · deploy

{
  "stack": { "target": "client-acme.default" },
  "gateway": { "routes": [{ "name": "site", "type": "prefix",
    "target": { "kind": "bucket", "name": "site" }, "default_key": "index.html", "fallback_mode": "not_found" }] },
  "resources": {
    "functions": [{ "name": "api" }],
    "buckets": [{ "name": "site" }],
    "tables": [{ "name": "content", "key_name": "page_id", "key_type": "S", "indexes": [] }]
  }
}

FAQ

Frequently asked questions

How do clients review work without a staging lecture?

They click a preview link posted into their bound conversation. The preview runs the same declared stack as production will, so “approved” means the thing itself, and the approval sits in the project record when scope is discussed later.

Can we see what each client costs us?

Yes — per-client projects carry per-client balances, covering build previews, agent demos, and running cost. Retainer and invoice conversations start from the meter instead of from estimates reconstructed quarterly.

What exactly changes hands at handoff?

The project: hosted git repository with code and stack.json, data in tables and buckets, and the conversation history. The client’s next team deploys with the same CLI the studio used — continuity is the deliverable.

Connected in the Project Graph

Related capabilities, architectures, and comparisons

Ship your stack on the Software Lifecycle Cloud.

Closed preview on dedicated European infrastructure. Bring a repository or start from a blueprint.

Talk to us See pricing & credits →