Collection

Let AI do useful work without giving it accidental authority

Design AI-assisted workflows so tools, permissions, evidence, approvals, testing and monitoring match the real task and limit the blast radius of a bad output.

15 elements · Explore in any order.

4 checklists · 6 principles · 5 protocols

Narrow your search

15 elements

Checklist

Give the agent only the tools this job needs

Every extra tool is another verb the system can accidentally conjugate.

Read the idea
Principle

Start an agent read-only when writing is not required

Reading first gives you evidence. Writing first gives you evidence plus cleanup.

Read the idea
Principle

Prefer a task-shaped tool to an open-ended one

A function named `create_invoice_draft` has fewer dangerous interpretations than `run_anything`.

Read the idea
Principle

Run agent actions in the user's authorization context

A helpful assistant should not quietly become a shared administrator.

Read the idea
Protocol

Put approval immediately before the high-impact action

Approval at the start of a long plan is not approval of the action the plan eventually invented.

Read the idea
Principle

Treat retrieved content as data, not as new authority

The page you asked the agent to read should not get to rewrite the job description.

Read the idea
Checklist

Keep secrets out of the system prompt

A prompt is a poor vault, even when nobody intends to show it.

Read the idea
Principle

Enforce permissions outside the model

A sentence in a prompt is guidance. An authorization check is a control.

Read the idea
Protocol

Validate model output for the system that will consume it

Readable text becomes a different risk when the next component treats it as code.

Read the idea
Checklist

Log what the agent did, not only what it said

The chat transcript is not the audit trail if the real effects happened elsewhere.

Read the idea
Protocol

Cross-check the fact before the agent acts on it

Confidence is a writing style. Evidence is a different object.

Read the idea
Protocol

Keep an eval set that can embarrass the agent

A demo asks whether the system can succeed once. An eval asks where it reliably fails.

Read the idea
Principle

Test the agent under conditions that resemble the real job

A perfect lab result can still be a poor rehearsal.

Read the idea
Protocol

Monitor the agent after deployment

Deployment is the first time the model meets all the mess your test set forgot.

Read the idea
Checklist

Bound how much damage one agent run can do

Automation turns one mistake into throughput unless you design a brake.

Read the idea