surehand
Articles / Reference / Page 2

The reference for governing AI agents, one question at a time., page 2

One question, answered in full, for the people who have to sign off on an agent.

posts
31
updated
feed
rss
Controls

How a hold works: triggers, named approvers, cover and time limits

Reference5 min read

Human in the loop only works if each hold has four parts: a trigger, one named approver, a named cover, and a time limit. What each part does, and what happens when one is missing.

Risks

Indirect prompt injection through documents: when the invoice gives the orders

Reference5 min read

Indirect prompt injection hides instructions in the documents your agent reads: invoices, claims, tickets, emails. What it is, why filters do not end it, and where the defence has to sit.

Regulation

ISO/IEC 42001 vs the NIST AI RMF, in plain English

Reference6 min read

One is a certifiable management-system standard. The other is a voluntary framework. What each asks of a company running AI agents, and what a certificate proves.

Controls

Kill switch, pause and rollback: what 'stop the agent' has to stop

Reference5 min read

A kill switch that stops new runs but not the one mid-payment is half a switch. What stop has to cover: in-flight actions, queued actions, scheduled runs and the credentials.

Security

Least privilege for AI agents: how to scope what an agent can touch

Reference5 min read

Most agents run on a borrowed login with far more access than the job needs. Scope them by action, not by system, and enforce it outside the model.

Vendors

Microsoft Copilot Studio approvals: what the human review actions do, and where they stop

Reference7 min read

A fair reading of Copilot Studio's Request for information action and multistage approvals, from Microsoft's docs: what they pause, who they ask, and what they leave to you.

Vendors

Model Context Protocol for operations teams: what to ask before you connect a server

Reference5 min read

MCP is the standard way agents plug into tools and data. A server can read your systems and act in them. What the spec itself says about consent and safety, and what to check before connecting one.

Vendors

n8n human in the loop: Wait nodes, tool approval, and the who-clicked problem

Reference7 min read

n8n's three ways to put a person in front of an agent's action, read from its docs, and the difference between a link anyone can click and a verified approver.

Security

Prompt injection in accounts payable: what an invoice can tell your agent to do

Reference5 min read

An invoice is text a stranger wrote, and your agent reads all of it. Prompt injection turns that text into instructions. You can't filter it out. You can limit what it can do.

Risks

Retries, idempotency and side effects: why a retried agent pays twice

Reference5 min read

An agent that retries 'pay invoice' after a timeout can pay it twice. What idempotency means, why agents make it worse, and what to check before an agent gets write access.

Controls

Segregation of duties for AI agents: when one agent initiates and approves

Reference5 min read

Segregation of duties says no one person should initiate, approve and record the same transaction. An agent that does all three breaks the rule. How to map SoD onto agents.

Operations

What is shadow mode for AI agents?

Reference5 min read

Shadow mode runs the agent on live work while your people still do the job. The agent decides, nothing happens, and you compare. It is the cheapest go-live test there is.

The product

The controls in these notes are what Gatehouse, Surehand’s control plane, enforces at runtime.

See Gatehouse
[ your next step ]

Bring us the queue nobody wants.

One process, studied in writing. You keep the document, whatever it says.

support@surehand.io