Netskope Extension Risk

Detail view · rubric v3.6
← Back to catalog

DeepSeek Chat Exporter

cohbpcihoiahgokbjkgkecodljploimg
Risk Score
4.73
Risk Level: Medium
Recommendation: 🟡 MEDIUM RISK — review
Category AI
Installs 10,000
Rating 4.5
Last updated 2025-11-16 (7 months ago)
Manifest version MV3
CSP present ❌ no
Developer lyqtzs@gmail.com
Verified publisher ✅ yes
Featured by Google ❌ no
Privacy policy link
Web Store open ↗

Top Risks

  • Brand impersonation: uses 'DeepSeek' name with no affiliation; developer is anonymous gmail user with no verified identity.
  • Privacy policy is Google's generic account policy — does not scope to this extension; admits data collection and 3rd-party sharing.
  • Content script runs on all DeepSeek chat pages and can access full conversation content.
  • install_url_hijack flag set: onInstalled opens a URL — potential unsolicited redirect on install.
  • No CSP; DOM-XSS sink (innerHTML) in bundled pdfobject.min.js with no mitigating policy.

Evidence

  • brand_impersonation store brand_mention.is_impersonation=true; confirmed_owner=false; developer_email is gmail with no business domain.
  • install_url_hijack crx install_url_hijack=true; target null but onInstalled redirect present — classic unwanted redirect pattern.
  • generic_privacy_policy api Policy URL is Google account policy; scope_extension=false, data_collection=true, third_party_sharing=true.
  • dom_xss_sink crx pdfobject.min.js: innerHTML assigned from variable; no CSP to mitigate; csp_present=false.
  • free_webmail_anonymous_dev store developer_name empty, developer_email lyqtzs@gmail.com, no business website, no verified publisher badge.
  • js_external_hosts crx Extension fingerprint includes deepseek-ai.online, github.com, pipwerks.mit-license.org.
  • content_scripts_on_target_domain manifest Scripts injected into *://chat.deepseek.com/* and *://deepseek.com/* — can read full chat conversations.
  • no_csp manifest content_security_policy is null; MV3 default is restrictive but dom sink with no explicit policy raises risk.

Permissions Breakdown

  • storage low Stores local settings/exports; minimal risk on its own.
  • content_scripts(*://chat.deepseek.com/*, *://deepseek.com/*) medium Script runs on DeepSeek chat pages; can read conversation content.

Pillar Scores

Permissions1.30
Reputation8.50
Network2.00
Webstore5.00
Maintenance1.50
Privacy10.00
Code Quality2.00
CVE Exposure0.00

Bookkeeping

Rubric v3.6
Scored at 2026-06-16 07:25
Listing SHA 1807ce57dde5…
Force block — not fired
Score recovered no
Elapsed 22.5s