Is ollama.com safe?

suspiciousmedium confidence
46/100

context safety score

A score of 46/100 indicates multiple risk signals were detected. This entity shows patterns commonly associated with malicious intent.

identity
100
behavior
80
content
17
graph
30

8 threat patterns detected

medium

encoded payload

suspicious base64-like blobs detected in page content

medium

malicious redirect

script/meta redirect patterns detected in page source

high

brand impersonation

The page impersonates Anthropic's Claude Code by rendering a fake terminal UI showing 'Claude Code v2.1.37' running on the qwen3 model. This deceives users into believing they are launching genuine Anthropic Claude Code, when the underlying model is qwen3 via Ollama. The fake version string and terminal chrome are designed to mimic Claude Code's real interface. (location: page.html lines 351-373, panel-coding section)

high

brand impersonation

'OpenClaw' is presented as an 'open source AI assistant' and promoted throughout the page with its own tab, terminal demo, and integration listing. The name closely mimics both 'OpenCode' (a real product listed on the same page) and Anthropic's 'Claude', blending the two to create a confusingly similar brand identity. The integration URL slug further differs ('clawdbot') from the display name 'OpenClaw', indicating deceptive naming. (location: page.html lines 323, 379-408, 520-522; integration link href='https://docs.ollama.com/integrations/clawdbot')

medium

brand impersonation

The page uses official brand logos and names of Anthropic (Claude Code), OpenAI (Codex), and OpenCode alongside fictional/unverified products (OpenClaw) without clear affiliation disclaimers, implying endorsement or official integration status that may not exist for all listed items. (location: page.html lines 488-496, integrations section)

medium

social engineering

The page promotes direct shell execution via 'curl -fsSL https://ollama.com/install.sh | sh' with a one-click copy button and instructional text 'paste this in terminal'. Embedding this alongside impersonated brand UIs (Claude Code, Codex) increases the likelihood a user will trust and execute the remote script, amplifying the social engineering risk. (location: page.html line 311-312, main hero section)

low

social engineering

The claim 'Over 40,000 integrations' is displayed prominently with no substantiation, used to establish false credibility and urgency for account creation. This is a classic social proof manipulation tactic combined with a call-to-action to 'Create account'. (location: page.html line 483, integrations section heading and sign-up CTA section lines 554-566)

medium

malicious redirect

The 'OpenClaw' automation integration card displays the label 'OpenClaw' but its href points to 'https://docs.ollama.com/integrations/clawdbot' — the slug 'clawdbot' does not match the displayed product name. This URL/label mismatch is a redirect deception pattern where the visible link text misleads users about the actual destination. (location: page.html line 520, Automation integrations section)

API

curl https://api.brin.sh/domain/ollama.com

FAQ: how to interpret this assessment

Common questions teams ask before deciding whether to use this domain in agent workflows.

Is ollama.com safe for AI agents to use?

ollama.com currently scores 46/100 with a suspicious verdict and medium confidence. The goal is to protect agents from high-risk context before they act on it. Treat this as a decision signal: higher scores suggest lower observed risk, while lower scores mean you should add review or block this domain.

How should I interpret the score and verdict?

Use the score as a policy threshold: 80–100 is safe, 50–79 is caution, 20–49 is suspicious, and 0–19 is dangerous. Teams often auto-allow safe, require human review for caution/suspicious, and block dangerous.

How does brin compute this domain score?

brin evaluates four dimensions: identity (source trust), behavior (runtime patterns), content (malicious instructions), and graph (relationship risk). Analysis runs in tiers: static signals, deterministic pattern checks, then AI semantic analysis when needed.

What do identity, behavior, content, and graph mean for this domain?

Identity checks source trust, behavior checks unusual runtime patterns, content checks for malicious instructions, and graph checks risky relationships to other entities. Looking at sub-scores helps you understand why an entity passed or failed.

Why does brin scan packages, repos, skills, MCP servers, pages, and commits?

brin performs risk assessments on external context before it reaches an AI agent. It scores that context for threats like prompt injection, hijacking, credential harvesting, and supply chain attacks, so teams can decide whether to block, review, or proceed safely.

Can I rely on a safe verdict as a full security guarantee?

No. A safe verdict means no significant risk signals were detected in this scan. It is not a formal guarantee; assessments are automated and point-in-time, so combine scores with your own controls and periodic re-checks.

When should I re-check before using an entity?

Re-check before high-impact actions such as installs, upgrades, connecting MCP servers, executing remote code, or granting secrets. Use the API in CI or runtime gates so decisions are based on the latest scan.

Learn more in threat detection docs, how scoring works, and the API overview.

Last Scanned

March 4, 2026

Verdict Scale

safe80–100
caution50–79
suspicious20–49
dangerous0–19

Disclaimer

Assessments are automated and may contain errors. Findings are risk indicators, not confirmed threats. This is a point-in-time assessment; security posture can change.

start scoring agent dependencies.

integrate brin in minutes — one GET request is all it takes. query the api, browse the registry, or download the full dataset.