Promptective journal / Topic index

AI security

5 journal entries tagged AI security.

A malicious repository directs an AI agent to read a sensitive file and make an outbound request

Article29 August 20265 min read

A malicious repository could turn Amazon Kiro into a data-exfiltration path

A Kiro prompt-injection flaw shows how repository content can steer an AI coding agent towards sensitive files, configuration changes and data exfiltration.

Read article

An Amp thread passing through a Promptective policy aperture and continuing along three protected paths

News29 August 20268 min read

The Hook Between Intent and Action

Promptective now supports Amp.

How Amp’s plugin architecture and Orb identity helped Promptective bring managed policy to direct prompts, tool calls and tool results.

Read news

A compact production model server stands inside nested cobalt, smoked-glass and aluminium control trays

Guide21 August 20265 min read

LLM security in production

Safer inputs, retrieval, outputs and tools.

A practical guide to LLM security in production: control prompts, retrieval, outputs and tools; minimise sensitive evidence; test and respond.

Read guide

A graphite agent gripper holds a brass tool above a cobalt approval boundary beside an already permitted low-risk tool

Guide2 February 20264 min read

Securing AI agents with tools

Bound permissions and hold consequential actions for approval.

Learn how to reduce prompt injection, tool abuse, and MCP risks with practical controls for LLM agents and their connected systems.

Read guide

A cream trusted instruction and a dark untrusted ribbon cross before entering one model chamber through a cobalt aperture

Guide15 January 20262 min read

What is prompt injection?

How it works and how to reduce the risk.

Learn how prompt injection attacks work, see practical examples, and compare controls for protecting LLM applications in this 2026 guide.

Read guide

All Articles, Guides and News