Agent with enforced write control architecture
The pattern for letting an agent touch production: every write passes an approval gate, every action lands in a hash-chained evidence log.
Every resource, and what it costs.
Projections from August 2026 list prices for always-on resources. Connect an account and these become the figures your provider actually bills.
| Node | Type | What it is | Projected |
|---|---|---|---|
| Production database | Database | db.t4g.large · multi-AZ · 100 GB | $215/mo |
| Evidence pack storage | Storage | 100 GB standard | $3.00/mo |
| Approval notifications | Messaging | 1M notifications | $1.00/mo |
| Task queued | Trigger | Internal API | — |
| Plan the change | Agent step | Decides what to write and why | billed per token |
| Approval gate | Guardrail | Every production write pauses here | enforced at runtime |
| db_write | Tool | Applies the approved change | — |
| Budget cap | Guardrail | Halts runs over $80/mo | enforced at runtime |
| ArcFoundry runtime | Runtime | — | — |
Agent steps are priced from provider-reported token usage once the agent runs, not estimated. Guardrails cost nothing and are the reason a runaway agent cannot.
Similar templates.
Support agent on live systems
A support agent that reads the orders database and writes back through an approval gate. Agent steps and the systems they touch on one graph.
Retrieval service (RAG)
An answer endpoint over your own documents: vector store, retrieval step, generation step, cached responses.
Document processing pipeline
Files land in storage, an agent extracts structured fields, a person approves anything low-confidence, and the result is written to the database.
Open Agent with enforced write control on the canvas.
It loads as an editable graph. Connect an account or instrument an agent and the projected figures above become measured ones.