MAQPNA

Learn

OWASP Top 10 for Agentic Applications mapped to controls.

ASI01 to ASI10, one row each: the risk, what helps, the MAQPNA control and where it falls short.

Updated 31 October 2026.

What is the OWASP Top 10 for Agentic Applications?

The OWASP Top 10 for Agentic Applications is a list of the ten most critical security risks of AI agents: systems that plan, call tools, keep memory and act with delegated credentials. The OWASP GenAI Security Project published the 2026 edition on 9 December 2025, with items numbered ASI01 to ASI10. It is a voluntary risk catalogue, not a standard you certify against.

It sits next to the OWASP Top 10 for LLM Applications 2025 and does not replace it. The LLM list covers the model and the application around it: prompt injection (LLM01), excessive agency (LLM06), unbounded consumption (LLM10) and so on. The agentic list covers what happens once the model can act. Most agent reviews need both. The Agentic Security Initiative page links the full document.

How does each ASI risk map to a runtime control?

Names are as OWASP publishes them. "Mitigates" means the root cause lies elsewhere, usually in the model. The same mapping, in short, is on the trust centre.

ASI id · riskWhat helpsMAQPNA controlCoverageDocs
ASI01 · Agent Goal HijackTreat retrieved content as untrusted; least privilege that holds whatever the model decides; gate exfiltration after untrusted inputTaint tracking holds or denies sinks (push, send, post) after a session reads untrusted content; injection guards on tool output; per-tool policyMitigates: contains the effect, the model can still be fooledTaint
ASI02 · Tool MisuseLeast agency: allow only the tools and arguments a task needs; rate limits; approval for destructive callsPolicy per tool and argument (typed conditions, Cedar, Rego), default deny, rate limits, budgets, approvals; policy test, replay and eval before rolloutEnforcedPolicies
ASI03 · Identity & Privilege AbuseA distinct, short-lived identity per agent; act on behalf of a named person; no shared keysSigned session token per session with an on-behalf-of claim, per-call scope checks, narrowing-only delegation, upstream credentials kept in the gateway or a vault, optional SVID mTLSEnforcedIdentity
ASI04 · Agentic Supply Chain VulnerabilitiesPin and verify tools, images and servers; keep an inventory of what agents actually useTool pinning (a changed tool definition is hidden and denied), registry allow-lists, digest-pinned images, cosign-verified releases, a CycloneDX AI-BOMEnforced for tools and images; mitigates for third-party modelsTool pinning
ASI05 · Unexpected Code ExecutionRun generated code in a sandbox with no route out except a checked oneA sandbox per session (gVisor, microVM or confidential VM), non-root, read-only root, no service-account token, default-deny network with the gateway as the only exitContainedSandboxes
ASI06 · Memory & Context PoisoningScope memory per agent and user; scan what is written; expire and eraseGoverned memory stores scoped per agent and user, DLP on writes, retention, erasure with a signed certificate; ephemeral sessionsMitigatesDLP and taint
ASI07 · Insecure Inter-Agent CommunicationAuthenticate peers; allow-list who may call whom; keep the delegation chainGoverned A2A: signed agent cards, peer allow-lists, nested delegation chain with depth limits, policy and audit on both hopsEnforced inside one installation; cross-organisation clearing is plannedMCP and A2A
ASI08 · Cascading FailuresLimit blast radius; hard spending and rate limits; fail closedSandbox and network limits, budgets with hard stops, rate limits, session TTLs, fail-closed defaults, model fallback with circuit breakersMitigatesBudgets
ASI09 · Human-Agent Trust ExploitationShow approvers what will really happen; separate duties; confirm with the affected userApprovals bound to the exact call, four-eyes quorum, no self-approval, redacted argument previews, confirmation by the user on their own device (CIBA)Enforced for held callsApprovals
ASI10 · Rogue AgentsA way to stop an agent at once, and a record to prove what it didKill switch across the installation within seconds, session timeline, posture checks, alerts, hash-chained audit ledgerStop and prove: enforced. Behavioural anomaly scoring: not built inKill switch

What does MAQPNA not cover?

A runtime limits and records what a fooled model can do. Some parts of the list need other controls.

GapRelated ASIWhyWhat helps instead
The model being persuadedASI01, ASI06No runtime can stop a model being fooled; MAQPNA contains the effectModel-side defences, evaluation and red-teaming
Behavioural anomaly detectionASI10Not built inA SIEM or UEBA reading the ledger stream
Weights and training data of third-party modelsASI04MAQPNA governs the route to a model, not what is inside itSupplier due diligence, model cards, your own evaluation
Wrong but allowed answersASI09An allowed call with a misleading output is a model-quality problemHuman review of outputs, grounding, user training
Agents outside the gatewayAllControls apply only to traffic that goes through itDefault-deny egress so the gateway is the only way out

Where MAQPNA fits

How do you show an auditor the OWASP controls work?

A mapping on a web page is a claim. Evidence is a record. MAQPNA builds an evidence pack for the owasp-asi framework from the tamper-evident audit ledger and the configuration in force: one system, one period, signed, with CSV workpapers of approvals, denials by reason, DLP blocks and revocations. Add the AI-BOM and the pack cites it under ASI04 and ASI06.

Each control is marked evidenced, partial or "not covered by MAQPNA (customer process)". A pack supports a review; it does not certify anything, and its mapping stays marked draft until counsel and audit partners review it. Anyone can verify a pack offline. Generating one needs a licence with the evidence-packs feature or the Sovereign edition.

For the controls behind the rows, read prompt injection in agent tool calls (ASI01), what an MCP gateway is (ASI02 to ASI04), human-in-the-loop approval (ASI09) and the kill switch (ASI10). The full threat model is in the security model. To see how it fits a wider programme, read AI agent governance, the EU AI Act for AI agents and ISO 42001 and the NIST AI RMF, or browse solutions by sector.

FAQ

Common questions

What are ASI01 to ASI10?

The ten items of the OWASP Top 10 for Agentic Applications (2026): Agent Goal Hijack, Tool Misuse, Identity & Privilege Abuse, Agentic Supply Chain Vulnerabilities, Unexpected Code Execution, Memory & Context Poisoning, Insecure Inter-Agent Communication, Cascading Failures, Human-Agent Trust Exploitation and Rogue Agents.

How is it different from the OWASP Top 10 for LLM Applications?

The LLM list (2025) covers models and LLM applications, such as prompt injection and excessive agency. The agentic list covers agents that act with tools, memory, credentials and other agents. It extends the LLM list rather than replacing it.

Does MAQPNA make an agent OWASP compliant?

No. The list is a voluntary risk catalogue, and no product makes you compliant with it. MAQPNA supplies runtime controls for each item and evidence that they ran, and it says where it only mitigates or does not cover a risk.

Can a runtime stop agent goal hijack?

It cannot stop the model being fooled. It can stop a fooled model doing damage: taint rules hold or deny exfiltration after untrusted input, and policy limits every tool call whatever the model decided.

Put a control behind every risk.

Run an agent through the gateway on your laptop today, then generate an OWASP evidence pack from the ledger.

More install options
curl -fsSL https://maqpna.com/install.sh | sh