Tag

#runtime-security

23 posts0 followers

Explore Hashnode

Alternatives

Rain S19h ago · 16 min read

Actions the Tool Registry Does Not Know About: Who Owns Them?

One application, one events table, two rows. One row comes from your AI assistant: it calls create_event, a registered tool marked R3 that waits for your confirmation. The other comes from POST /api/v

rain6fish.hashnode.dev

Rain S5d ago · 9 min read

An Auditor Is Not an Admin: Giving Someone the Evidence Is Not Giving Them the System

An auditor has to review what an AI did. The cheapest way to let them do that is to hand them an admin account. But then the person they are supposed to review becomes themselves. This piece is about

rain6fish.hashnode.dev

Rain S5d ago · 9 min read

Silence Reads as Completeness: An AI Audit Package Must Carry a Coverage Statement

An auditor receives an evidence package. The hash chain verifies. What will they read that as? Most people read it as two things: the rows were not altered, and they are all here. The chain proves onl

rain6fish.hashnode.dev

Rain S5d ago · 7 min read

One Timeout Cannot Answer Two Questions: Why an AI Write's Confirmation Needs Two Windows

An AI asks to write something, and a person has to approve it. There are two kinds of "expired" in that sentence, and they are not the same thing: should the conversation keep waiting for the approval

rain6fish.hashnode.dev

Rain S5d ago · 7 min read

A Unique Constraint Protects the Ledger, Not the Action: Making an AI Write Happen Once

An assistant calls a tool that writes. That call can arrive twice: the model retries it, the same confirmation gets decided twice, or the process restarts mid-flight. A second arrival must not write a

rain6fish.hashnode.dev

Rain S5d ago · 6 min read

Human Confirmation Cannot Be a Boolean: What an AI Write Has to Bind

The earlier pieces were about stopping things: where the boundary sits, how risk gets tiered, how a write gets undone. This one steps back and looks at a single part. What the system actually holds af

rain6fish.hashnode.dev

Rain S5d ago · 8 min read

When a Record Has No Word for "Unknown", an Empty Value Will Say "Everything Is Fine"

Something happened recently that I want to start with. We fixed a defect: four different kinds of "nothing" were collapsed into a single switch, so the readings couldn't tell "by design" from "broken"

rain6fish.hashnode.dev

Rain S5d ago · 4 min read

What Your Boundary Actually Covers: Enumeration Comes Before Strength

After the first few pieces went out, three questions came back. They don't look related: Is tool the right unit for an AI's actions? How do you monitor a path you didn't know existed? What about a slu

rain6fish.hashnode.dev

Rain S5d ago · 9 min read

Revocation Is Not a Button: How to Actually Undo an AI Side Effect

The first two pieces were about stopping things — put the boundary on the execution path, then give every tool an execution policy by risk. But some of it can't be stopped. The policy allowed it, the

rain6fish.hashnode.dev

Rain S5d ago · 5 min read

Why AI Writes Need Risk Tiers: The R0-R5 Tool Risk Model

The previous piece, "Runtime over Prompt", argued that the security boundary belongs on the execution path — immediately before a tool call can produce a side effect. This one goes a level deeper. Onc

rain6fish.hashnode.dev

Search Hashnode

Explore Hashnode

Trending tags this week

Actions the Tool Registry Does Not Know About: Who Owns Them?

An Auditor Is Not an Admin: Giving Someone the Evidence Is Not Giving Them the System

Silence Reads as Completeness: An AI Audit Package Must Carry a Coverage Statement

One Timeout Cannot Answer Two Questions: Why an AI Write's Confirmation Needs Two Windows

A Unique Constraint Protects the Ledger, Not the Action: Making an AI Write Happen Once

Human Confirmation Cannot Be a Boolean: What an AI Write Has to Bind

When a Record Has No Word for "Unknown", an Empty Value Will Say "Everything Is Fine"

What Your Boundary Actually Covers: Enumeration Comes Before Strength

Revocation Is Not a Button: How to Actually Undo an AI Side Effect

Why AI Writes Need Risk Tiers: The R0-R5 Tool Risk Model