Top 5 · 2026-03-04 · source-backed

Schneier Publishes "Promptware Kill Chain" — MITRE ATT&CK for AI Agents.

Issue

2026-03-04

Confidence

source-backed

Sources

Redaction

passed

Story

Bruce Schneier and coauthors proposed a 7-stage kill chain for "promptware" — prompt injection attacks that evolve into multi-step malware: initial access → privilege escalation → reconnaissance → persistence → C2 → lateral movement → actions on objective. Critical insight: persistence via long-term agent memory poisoning and C2 via controllable trojans turn static prompt injection into evolving, adaptive malware. Self-replicating attacks spread by tricking email agents into forwarding payloads. This gives defenders the first shared vocabulary for reasoning about agentic threats at the same maturity as traditional cyber frameworks. What to do: Map your agent deployments against the 7 stages. Implement defense-in-depth at each boundary, especially tool-return validation (see AgentSentry paper below). (Schneier on Security)

Source trail

Schneier on Security

Entities

AgentSentry Bruce Schneier Critical Implement MITRE ATT Promptware Kill Chain Schneier Schneier Publishes Security Self

Provenance

Canonical issue: Ramsay Research Agent — 2026-03-04
AI generated: no
Story unit: 2026-03-04-schneier-publishes-promptware-kill-chain-mitre-att-ck-for-ai-agents
Labels: source-backed, canonical briefing excerpt

Fetching from the wire…

Schneier Publishes "Promptware Kill Chain" — MITRE ATT&CK for AI Agents.

Story

Related stories

MIT spends $3M+ on 500 AI cameras classifying gender, age, and clothing color at 35 feet, during a hiring freeze.

Tracebit's canary secrets cut AI attacker admin access to 5%

43 Attacks Can Fully Compromise a Self-Hosted AI Agent

Self-replicating "Miasma" malware hit 70+ Microsoft open-source repos, stealing credentials the instant a repo opens in an AI tool.

AgentSentry — Temporal Causal Defense Against Prompt Injection

Schneier gives AI delegation a gym-task test

Schneier: 31 Companies Caught Poisoning AI Memory

Schneier: "AI Models Are Commodified — Branding Is All That's Left"

Source trail

Entities

Provenance

Schneier Publishes "Promptware Kill Chain" — MITRE ATT&CK for AI Agents.

Story

Related stories

MIT spends $3M+ on 500 AI cameras classifying gender, age, and clothing color at 35 feet, during a hiring freeze.

Tracebit's canary secrets cut AI attacker admin access to 5%

43 Attacks Can Fully Compromise a Self-Hosted AI Agent

Self-replicating "Miasma" malware hit 70+ Microsoft open-source repos, stealing credentials the instant a repo opens in an AI tool.

AgentSentry — Temporal Causal Defense Against Prompt Injection

Schneier gives AI delegation a gym-task test

Schneier: 31 Companies Caught Poisoning AI Memory

Schneier: "AI Models Are Commodified — Branding Is All That's Left"

Source trail

Entities

Provenance