context safety score
A score of 38/100 indicates multiple risk signals were detected. This entity shows patterns commonly associated with malicious intent.
phishing
4 deceptive links where visible host does not match destination host
hidden content
2 hidden or tiny iframe elements detected
malicious redirect
Third-party bot detection script loaded from cdn.perfdrive.com (ShieldSquare/Radware) with a beacon URL pointing to eupulse.shieldsquare.net. The inline script on line 26-27 contains invalid JavaScript syntax (bare URLs without quotes: `a.src = u;` where u is `https://cdn.perfdrive.com/aperture/aperture.js` and `ssConf("c1", https://eupulse.shieldsquare.net)`) indicating the rendered source may be tampered or malformed, and all user traffic fingerprints are being exfiltrated to an external third-party domain outside impots.gouv.fr. (location: page.html:16-27)
malicious redirect
Internal help link for FranceConnect troubleshooting points to `http://portus.iia.dgfip/portail/contacts` — a plaintext HTTP internal hostname (`portus.iia.dgfip`) that is not a public government domain and would not resolve for external users, potentially redirecting users to an unexpected or internal-only endpoint. A second identical pattern appears in the FranceConnect help accordion at line 898. (location: page.html:867,898)
hidden content
Mobile app banner (`#banniereSmart`) links to `https://play.test.google.com/store/apps/details?id=fr.gouv.finances.smartphone.android` — note the subdomain `play.test.google.com` instead of the legitimate `play.google.com`. The `test` subdomain is anomalous and may redirect to a non-Google controlled resource or a test/staging environment, potentially serving a malicious APK. (location: page.html:97-99)
hidden content
Button label in OTP resend section reads 'Renvoyer le code de sécuritéZZZ' — the trailing 'ZZZ' suffix is a debug/test artifact visible in rendered page text, indicating development residue that should not be in production and suggests the page may not be the genuine production version. (location: page.html:643, page-text.txt:458)
hidden content
SMS validation modal submit button reads 'Recevoir le lien de renouvellementTEST' — the trailing 'TEST' suffix is another debug artifact in production-rendered content, further suggesting this may be a staging or tampered copy of the page. (location: page.html:1834, page-text.txt:1164)
hidden content
Multiple modal dialogs contain placeholder text 'TITRE' as the literal heading content (e.g., checkSMSMod modal h2, confirmationSpi modal, confirmationPassword modal), indicating incomplete template substitution. This is consistent with a cloned or scraped page where dynamic title injection did not execute. (location: page.html:1749,1877,1927)
credential harvesting
The page collects highly sensitive French tax authority credentials including: 13-digit numéro fiscal (tax ID number), password, 7-digit online access number (numéro d'accès en ligne), revenu fiscal de référence (reference taxable income), date of birth, and 6-digit OTP/unlock codes — all in a single authentication flow. While the domain is legitimate (cfspart.impots.gouv.fr, 39-year-old domain, valid OV TLS cert), the presence of test/debug artifacts (ZZZ, TEST, TITRE placeholders) in production raises the possibility this is a cloned credential-harvesting page mimicking the real DGFiP portal. (location: page.html:259-278,583-585)
prompt injection
Base64-encoded value in `__uzdbm_2` JavaScript variable decodes to a session/tracking token containing what appears to be a UUID and an IP address (`fc2177570-d42e-53d8-5352-01030b749a1f$34.96.45.112`). While not a direct prompt injection, this obfuscated variable exfiltrates client fingerprinting data to the ShieldSquare CDN, and its presence could be used to track and manipulate AI agent sessions that render or process this page. (location: page.html:16)
curl https://api.brin.sh/domain/cfspart.impots.gouv.frCommon questions teams ask before deciding whether to use this domain in agent workflows.
cfspart.impots.gouv.fr currently scores 38/100 with a suspicious verdict and low 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.
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.
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.
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.
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.
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.
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.
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.
integrate brin in minutes — one GET request is all it takes. query the api, browse the registry, or download the full dataset.