Join our Newsletter — 33% off our NHI Course
Threats, Abuse & Incident Response

Slopsquatting

← Back to Glossary
By NHI Mgmt Group Updated October 31, 2026 Domain: Threats, Abuse & Incident Response

Slopsquatting is a supply-chain attack that exploits hallucinated package names suggested by AI systems. An attacker registers the invented name in a public registry and waits for a developer or build pipeline to install it. The risk sits at the intersection of model error, dependency trust, and software delivery speed.

How Slopsquatting Works

Slopsquatting begins when an AI system invents a plausible-sounding package name that does not exist, but looks legitimate enough to be trusted. The attacker then registers that name in a public package registry and relies on a developer, script, or build job to install it before anyone notices the name was hallucinated.

This makes the attack especially effective in fast-moving delivery pipelines, where dependency decisions are automated or only lightly reviewed. The danger is not just the bad package itself, but the false confidence created by a model-generated suggestion that feels discoverable and specific.

Why Slopsquatting Is Dangerous

Slopsquatting turns model error into a supply-chain entry point. If the hallucinated dependency is pulled into a project, the attacker can gain a path into source code, CI pipelines, developer machines, or downstream builds, depending on where the package is installed and what it can execute during install or runtime.

The attack is effective because package ecosystems reward speed and trust. Developers often assume that a named dependency has been vetted somewhere upstream, and AI-generated suggestions can blur the line between a real library, a typo, and a complete fabrication.

As NHI Management Group’s Malicious Package and Dependency Confusion Guide explains, package-namespace abuse, install scripts, provenance, and registry trust are all part of the same threat surface.

How Organisations Check Exposure

Exposure review starts with asking where AI is allowed to suggest or introduce dependencies, then checking whether those suggestions are automatically accepted by developers, copilots, dependency bots, or build systems. The key question is not only whether a package name appears in code, but whether any workflow can reach a registry and install it without human verification.

Teams should also look for internet-exposed registries, permissive dependency update rules, and projects that routinely accept new packages from AI-generated code or chat-based recommendations. A small number of uncontrolled package entry points can become a broad supply-chain problem if they are shared across many repositories or pipelines.

The issue is closely related to AI Coding Agents Security Guide, because coding assistants and terminal agents can turn a hallucinated suggestion into an actual dependency action.

Defences, Detection, and Response

Good defence combines dependency discipline with AI-specific guardrails. The strongest pattern is to require verification before installation, especially for new packages, generated manifests, and dependencies introduced by AI-written code. Provenance checks, package allow-lists, and restricted install behaviour reduce the chance that a hallucinated name becomes a real compromise path.

Detection should focus on unusual dependency introductions, install-time execution, and package names that resemble AI-generated suggestions or recent registry registrations. If a slopsquatted package is discovered after installation, responders should treat it as a potential supply-chain compromise, not just a harmless typo.

That is why the Slopsquatting and Package Hallucination Guide is useful for defensive detail, while the OWASP LLM Top 10 LLM09 Misinformation Guide helps frame the broader problem of trusting AI output without validation.

Practical Meaning For Secure Delivery

Slopsquatting is not just an AI curiosity. It is a delivery-control problem that sits between model output, package trust, and release velocity. The more an organisation lets AI influence dependency choice, the more it needs explicit controls around package verification, source provenance, and approval flow.

For teams building with AI coding tools, the most important shift is cultural as well as technical: treat dependency names from an AI system as untrusted suggestions until they are independently validated. In fast-moving pipelines, that assumption matters more than the sophistication of the model that produced the name.

For a broader control baseline, see AI Code Security and Agentic Coding Security Buyer's Guide, which covers package checks, sandboxing, and evaluation of coding-agent controls.

Standards & Framework Alignment

This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.

OWASP API Security Top 10 addresses the attack and risk surface, while CIS Controls v8, SLSA, OWASP SAMM and OWASP ASVS set the governance and control requirements practitioners need to meet.

FrameworkControl / ReferenceRelevance
OWASP API Security Top 10API9 — Improper Inventory ManagementSlopsquatting exploits unmanaged dependency inventory and unverified package entry points.
Recommendation — Inventory and validate every introduced dependency before it reaches build or deployment flows.
CIS Controls v8CIS-2 — Inventory and Control of Software AssetsPackage hallucinations become dangerous when software assets and dependencies are not controlled.
Recommendation — Track approved packages and block unvetted dependencies from entering development workflows.
SLSASupply-chain Levels for Software ArtifactsSlopsquatting is a software supply-chain integrity problem involving untrusted artifacts and provenance gaps.
Recommendation — Require stronger provenance and integrity checks before accepting externally sourced packages.
OWASP SAMMSoftware Assurance Maturity ModelThe term directly affects secure software delivery practices and dependency governance maturity.
Recommendation — Build dependency verification and secure release controls into the software delivery lifecycle.
OWASP ASVSV15 — Secure Coding and ArchitectureAI-generated dependency choices affect secure design and the trustworthiness of application components.
Recommendation — Review generated code paths for unsafe dependency introduction and enforce secure architecture gates.

Practitioner Guidance

Why practitioners should care: Slopsquatting converts a model hallucination into a real supply-chain dependency, so ownership should sit with both application security and platform engineering. The control objective is to stop unverified package names from becoming build inputs.

Common misunderstanding: Teams often assume a package is safe if it came from an AI assistant or was mentioned in a code suggestion. In practice, the suggestion itself is the risk signal, because the package may never have existed until an attacker registered it.

Practitioner takeaway: Treat AI-suggested dependencies as untrusted until they pass the same verification gates you would apply to any other external software component.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Reviewed and updated by the NHIMG editorial team on October 31, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org