Chat Highlight: Auto ChatGPT & AI Highlighter
ffnhgclfemimnnbkbhebjbobiipgkgbk
Risk Score
4.19
Risk Level:
Medium
Recommendation:
🟡 MEDIUM RISK — review
Top Risks
- <all_urls> + scripting + content_scripts on all pages: can read/modify any website the user visits.
- Brand impersonation: claims ChatGPT association but developer is unverified gmail user with no dev name.
- Free-webmail dev (iresearch666@gmail.com) with no developer name listed raises accountability concerns.
- Three innerHTML-from-variable DOM-XSS sinks across dashboard.js, popup.js, sidebar.js; no CSP mitigates.
- Supabase endpoint (othnrgszxxnqgdclatif.supabase.co) and Netlify host suggest third-party data routing for 57-install extension.
Evidence
- broad_host_access manifest <all_urls> in host_permissions and content_scripts_matches with scripting permission — reads/modifies all pages.
- brand_impersonation store brand_mention.is_impersonation=true for 'chatgpt'; developer_domain=gmail.com, confirmed_owner=false.
- free_webmail_no_devname store Developer email iresearch666@gmail.com; developer_name is empty string; no business identity.
- dom_xss_sinks crx dom_sink_innerhtml_userctrl in dashboard.js, popup.js, sidebar.js; no CSP present (csp_present=false).
- no_csp_mv3 manifest content_security_policy=null; MV3 default applies but no explicit CSP hardens innerHTML sinks.
- supabase_external_host crx othnrgszxxnqgdclatif.supabase.co in js_external_hosts — third-party DB endpoint for a 57-install AI extension.
- small_install_high_perm api install_perm_anomaly: 57 installs with high-tier permissions (<all_urls>, scripting) — tail attack surface.
- privacy_policy_adequate store Policy fetched; scope_extension=true, data_collection=true, retention=true, third_party_sharing=true — fully scoped.
Permissions Breakdown
- sidePanel low UI surface only, no data access.
- contextMenus low Adds right-click menu items; low standalone risk.
- storage low Local extension storage; limited scope.
- activeTab medium Grants access to current tab on user gesture; combined with scripting raises risk.
- scripting medium Programmatic script injection; paired with <all_urls> elevates to significant capability.
- identity low OAuth token access; risk depends on scopes requested at runtime.
- https://*.firebase.google.com/* low Google-owned backend for cloud sync; expected for stated function.
- https://*.firebaseio.com/* low Firebase Realtime DB; expected for cloud sync feature.
- https://*.googleapis.com/* low Google APIs; broad but Google-owned.
- https://firebase.googleapis.com/* low Specific Firebase endpoint; expected.
- https://*.firebaseapp.com/* low Firebase hosting; expected for this stack.
- https://gemini.google.com/* low Scoped to stated AI platform integration.
- <all_urls> high Broad host access combined with scripting and content_scripts — can read/modify any page.
Pillar Scores
Permissions5.50
Reputation7.00
Network3.50
Webstore5.50
Maintenance0.00
Privacy0.00
Code Quality3.50
CVE Exposure0.00
Bookkeeping
Rubric v3.6
Scored at 2026-07-08 12:53
Listing SHA
59bf15e13927…
Force block
— not fired
Score recovered
no
Elapsed
—