ASI05 · OWASP Agentic Top 10
In an agentic system, code generation and code execution happen in the same turn: the model emits an instruction and a tool runs it, with no human review step between. Attackers exploit this by injecting execution payloads into the agent's inputs; the realistic defence is at the runtime boundary (sandboxing, capability restriction, egress control), not at the generation step.
A classic web app generates code that a human reviews before it runs. An agent generates code and runs it in the same turn. The model emits an instruction (a shell command, a SQL statement, a Python snippet) and a tool executes it. The only thing between generation and impact is whatever sandbox the executing tool runs in.
Filtering generation is hard. LLMs are flexible enough that input sanitisation cannot reliably catch every malicious shape. The realistic containment is at execution: capability-restricted sandboxes, ephemeral filesystems, blocked outbound network, per-tool execution budgets. Assume the agent will eventually generate something bad and design the runtime so the blast radius is bounded.
Base-catalog T-numbers follow OWASP source material; normalized MAS scenario entries are Helmwart editorial cross-references. Role colour-codes Helmwart's display weight: chips in the hero use the same scheme.
An agent with code-execution or function-calling ability is steered, often through plain natural language, into running attacker-influenced code that the agent itself produces.
Open threat detail →An attacker exploits a security flaw in the agent framework software itself to run arbitrary code inside the agent's process.
Open threat detail →MITRE ATLAS catalogues adversary techniques against AI systems. The technique(s) below represent the red-team pivot for this entry: what an attacker is actually doing on the wire. Source: mitre-atlas/atlas-data v2026.09.
© 2026 The MITRE Corporation. ATLAS content is reproduced and distributed with the permission of The MITRE Corporation.
AML.T0102
Generate Malicious Commands
view on ATLAS ↗
Adversary uses an LLM to dynamically generate malicious commands from natural language, producing attack signatures that vary across executions.
Agentic angle: Agents with code-execution tools can be prompted to generate and immediately run adversary-crafted commands, collapsing generation and execution into one step.
From OWASP Appendix A (canonical inheritance)
OWASP AIVSS v0.8 is a companion OWASP methodology, published by the same project family as this Top 10, that extends a CVSS v4.0 base score with 10 Agentic Risk Amplification Factors — traits like autonomy, tool reach, and persistent memory that change how much damage a given vulnerability can actually do once an agent, not a human, is the one exploiting it. It scores individual findings, not ASI categories: when a threat mapped to ASI05 above ( T11, T20 ) fires on the canvas, its finding card shows an AIVSS score next to the risk band, built from the actual graph — autonomy level, tool privilege, memory access, and so on — rather than a generic per-ASI number.
Scoring happens per finding on the canvas, so it isn't reproduced on this reference page. Model a system that surfaces ASI05 to see it.
No single control answers an ASI; it is met by a layered stack. The cards below are ranked by how directly each control counters ASI05: the chips on each card name the threat of this ASI it actually covers, colour-coded by that threat's role.
Counters the core
Cover one or more of this ASI's primary threats — the strongest direct response.
An AI coding agent produces code that can be executed or merged to a production branch without a human ever reading it. If the agent has been manipulated, its generated code can contain hidden payloads, backdoors, or privilege-escalating logic. A code-generation review gate prevents that: every change attributable to an AI agent must pass automated static analysis and receive explicit human approval before it can merge or execute, and the agent identity that authored the change is structurally barred from also approving it.
When an agent executes generated or retrieved code, that code runs as a process with access to the host kernel. A vulnerability in the generated code, or a deliberate exploit injected through the agent's prompt, can reach the kernel and affect other workloads or the host itself. A kernel-isolated sandbox removes that path by giving the workload a kernel of its own: gVisor interposes a user-space kernel so syscalls reach the Sentry rather than the host, and Kata Containers runs the pod in a lightweight VM with a separate kernel. Which one is available is a property of the platform, not of the control — AKS ships Kata as Pod Sandboxing and does not support gVisor; GKE ships gVisor.
An AI agent operating with broad authority can propose actions that are irreversible: deleting records, modifying IAM policies, moving funds. A single human reviewer at the approval gate is a single point of failure, one compromised account, one fatigued reviewer, or one successful social-engineering attempt is enough to commit the action. Human dual-control addresses that by requiring two distinct, independent humans to approve before the action commits.
Agentic systems can act faster than a human can intervene through normal channels. A kill switch is the operational guarantee that a named human role can stop agent activity at any scope (single instance, class, or global) through a documented runbook, without requiring a code change or redeployment, and with every invocation written to an audit trail.
Content-layer inspection reads outbound payloads and decides pass, redact, or quarantine per channel — but it can only classify what it can parse as content. A direct TCP/UDP socket opened by agent-executed code, a raw DNS query, or a payload encoded into DNS subdomain labels never reaches an HTTP-shaped inspection point at all. Network egress isolation closes that gap one layer down: every agent and tool-bus process sits behind a default-deny network policy, and the only path to the internet is through an enforced proxy that mediates, allowlists, and logs each connection attempt regardless of what it contains.
An agent produces code, configuration files, tool-call payloads, and log records continuously and at a rate no human reviewer can match. Any of those artefacts may contain a live API key, service token, or private certificate, placed there accidentally through model context, or deliberately through prompt injection or context poisoning. Secret scanning places an inspection gate at every agent output seam: regex patterns match known token formats, entropy analysis detects arbitrary high-entropy strings, and validator calls confirm which candidates are live credentials. The CI-secret-scanning pattern is mature; the agentic specialisation is seam placement, moving the scanner from the repository gate to the agent egress point, where artefacts can be intercepted before they reach any downstream system.
An AI agent assembles itself at runtime from a model, prompt templates, plugins, and library dependencies, any of which can be tampered with before they arrive. A signed AI Bill of Materials (AIBOM) locks down that assembly: it records every component with a version and hash at build time, signs the manifest, and verifies it before the agent accepts traffic. A component that does not match its declared hash cannot silently enter the agent.
An agent that can generate and execute code treats code generation as a tool call and code execution as the outcome. If the generated code contains a known-dangerous pattern, no amount of prompt engineering stops it from running once the execute call goes through. Static analysis closes that gap: it scans every code artifact the agent emits against a rule set before execution is permitted, catching the vulnerability patterns the same tooling already catches in human-written code.
Press play. The voice downloads once, then it’s cached.
On-device · private