Guide

The OWASP Top 10 for agentic applications, control by control

OWASP published its first list of risks for AI agents in December 2025. This is what each entry looks like in practice, and the control that answers it.

By the NoFly team

The list

The OWASP Top 10 for Agentic Applications for 2026 was released on 9 December 2025 by OWASP’s Agentic Security Initiative. It covers systems that plan, call tools, keep memory and hand work to each other, rather than models that only answer questions.

Each risk, and the control for it

  1. ASI01

    Agent Goal Hijack

    An attacker redirects what the agent is trying to do, usually through content it reads.

    ProtectExplain

    Content from outside is screened before the agent acts on it, and the cause of a risky action is traced to the passage behind it.

  2. ASI02

    Tool Misuse and Exploitation

    Legitimate tools are used in harmful ways through injection, misalignment or unsafe delegation.

    Protect

    Every tool call is decided before it runs against a per-agent allow-list, and risky writes wait for a person.

  3. ASI03

    Identity and Privilege Abuse

    Delegation chains and inherited credentials are used to reach more than the agent should.

    ProtectGovern

    Each agent has its own identity and short-lived brokered keys, hand-offs cannot widen authority, and unused grants are surfaced for removal.

  4. ASI04

    Agentic Supply Chain Vulnerabilities

    Models, tools, MCP servers or other agent components are malicious or tampered with.

    DiscoverInterrogate

    MCP servers, skills, model files and packages are scanned before they run and matched against advisories, and live servers are questioned for a second face.

  5. ASI05

    Unexpected Code Execution

    Generated code or tool access turns into code running where it should not.

    Protect

    Shell commands are judged in full before they run, and critical moves such as a script piped from the internet into a shell are refused on the machine.

  6. ASI06

    Memory and Context Poisoning

    Corrupted memory or retrieved context steers what the agent does later.

    ProtectDiscover

    Poisoned memory is quarantined before it is read, and agent memory files and retrieval sources are scanned.

  7. ASI07

    Insecure Inter-Agent Communication

    Messages between agents lack authentication, integrity or validation.

    ProtectExplain

    Hand-offs are checked for a proven sender and for widened authority, and authority can only narrow from one hop to the next.

  8. ASI08

    Cascading Failures

    One fault spreads across agents into harm across the system.

    ProtectGovern

    Runaway loops are caught, spend is capped per agent, and automations can suspend an agent or clamp a budget the moment a threshold is crossed.

  9. ASI09

    Human-Agent Trust Exploitation

    People’s trust in the agent is used to get harmful actions approved.

    Govern

    Approvals show the real call rather than the agent’s summary, are bound to that exact request and expire as a no.

  10. ASI10

    Rogue Agents

    Agents drift from their intended function or scope.

    ProtectDiscover

    Each agent has a behavior baseline, a first-time tool or destination raises a flag, and any agent can be suspended or revoked in one step.

Using the list

Treat the list as a coverage map, not a checklist to tick once. For each entry, ask which control answers it on every surface your agents use, laptops, gateways, hosted platforms and your own code, then attack that control to see whether it holds.

How NoFly helps

  • Govern

    Every rule, finding and control maps to the OWASP Agentic Top 10, alongside the OWASP LLM Top 10, MITRE ATLAS and NIST AI RMF, with controls marked covered, attested, at risk or not assessed.

  • Attack

    Attack scenarios are mapped to the same list, so you can see which entries were tested, which held and which have not been tested yet.

  • Verify

    Controls, including other vendors’, are graded by what they did under attack, so coverage means proven rather than claimed.

What to do this week

This week3 actions
  1. For each of the ten entries, name the control that answers it today.
  2. Mark the entries where no control exists, or where it has never been tested.
  3. Pick the highest-risk gap and attack it this month.
Keep reading

More from the blog.

Find out what your AI is allowed to do today.

Book a call and we will map your agents, MCP servers and keys with you, or start on one laptop with the free scan.

Book a demo
npm i -g @nofly/agent