Every agent acts on someone's authority. These guides show where Smartflow sits for each platform and each function, which actions pause for a person with the authority to approve them, and what the record shows afterward. Each guide is a short read on the page, with a longer PDF for the team running the rollout.
By platform
Smartflow for Claude Code Model calls, MCP tools and production changes, governed across an engineering organization.
Smartflow for Claude Cowork Both deployment modes, and approvals by function in place of a read-only rollout.
Smartflow for OpenClaw Hard spend caps, keys that stay on the machine, and company policy for employees who run it.
Smartflow for Hermes Agent Budgets for scheduled jobs, command approvals, and a record of every call.
By function
Smartflow for IT Privileged access, configuration and deploys, with the approver on the record.
Smartflow for HR Employee records and people decisions, with the review and retention the new rules require.
Smartflow for Payroll Direct deposit changes and payments, held for approval before money moves.
How every guide is built
Each guide answers the same four questions.
- What can the agent reach?
- Where does Smartflow sit in the path?
- Which actions pause for a person?
- What does the record show?
The answers change by platform and by function. The pattern does not. The agent does the work. A person with authority approves the actions that matter. The record proves who decided.
Which guide do you need next?
Tell us the platform or function your teams are rolling out, and we will cover it. Talk to us.
Put this in the path of your own agents.
Policy enforced inline between your agents and every model and tool they reach, with a record bound to the human who owns it.
Request a Demo Read the docs