Netskope Extension Risk

Detail view · rubric v3.6
← Back to catalog

Prompt Optimizer - SecondBrain

aajjgdpofhhcjmjoombjdfepplndhgcp
Risk Score
6.07
Risk Level: High
Recommendation: 🟠 HIGH RISK — review
Category AI
Installs
Rating
Last updated
Manifest version MV3
CSP present ✅ yes
Developer
Verified publisher ❌ no
Featured by Google ❌ no
Privacy policy link
Web Store open ↗

Top Risks

  • Privacy policy is Google's generic account policy — does not scope to this extension, admits data collection and 3rd-party sharing (Privacy: 10.0).
  • webRequest + https://*/* enables full request interception across all sites; AI content processed by secondbrain.is backend.
  • Uninstall URL hijack flag set; post-uninstall redirect to unknown destination is a monetization/tracking indicator.
  • No developer name, email, install count, or update date — identity and maintenance status unverifiable.
  • innerHTML from variable in content script (DOM-XSS sink) running on all HTTPS pages including AI chatbots.

Evidence

  • privacy_policy_generic_google store Privacy URL points to myaccount.google.com/privacypolicy — Google's own policy, not scoped to this extension; data_collection+third_party_sharing=true.
  • uninstall_url_hijack crx chrome.runtime.setUninstallURL() called; target not captured but constitutes post-uninstall tracking signal.
  • broad_host_plus_webrequest manifest webRequest + https://*/* allows observation of all HTTPS traffic across every site the user visits.
  • external_js_hosts crx JS contacts optimize.secondbrain.is and secondbrain.is — prompt data likely sent to developer backend.
  • dom_xss_sink crx innerHTML assigned from variable in content/ui.js running on all HTTPS pages; DOM-XSS risk.
  • no_developer_identity store Developer name, email, install count, and last_updated all missing — accountability unverifiable.
  • content_scripts_broad manifest Content scripts match http://*/* and https://*/* — injected into every page, not just AI chatbots.
  • ai_extension_page_content manifest Extension processes prompts across ChatGPT, Claude, Gemini, Grok, Perplexity, Meta AI, Copilot — wide AI content reach.

Permissions Breakdown

  • webRequest high Can intercept/observe all HTTPS network requests; paired with broad host access.
  • https://*/* high Broad host access covering all HTTPS sites amplifies every other permission.
  • tabs medium Access to tab URLs, titles, and navigation events across all open tabs.
  • storage low Local data persistence; low risk in isolation.
  • unlimitedStorage low Removes storage quota; low risk but enables large local data accumulation.
  • alarms low Scheduling; minimal standalone risk.
  • sidePanel low UI surface only; no data access by itself.
  • background low Persistent service worker; extends capability persistence.

Pillar Scores

Permissions6.50
Reputation7.00
Network4.00
Webstore5.50
Maintenance5.00
Privacy10.00
Code Quality2.00
CVE Exposure0.00

Bookkeeping

Rubric v3.6
Scored at 2026-09-01 08:14
Listing SHA 89dcc9d34f22…
Force block — not fired
Score recovered no
Elapsed