Sider: Chat with all AI: GPT-5, Claude, DeepSeek, Gemini, Grok
difoiogjjojoaoomphldepapgpbgkhkb
Risk Score
7.00
Risk Level:
High
Recommendation:
🟠 HIGH RISK — review
Top Risks
- THREAT FEED: extension ID independently confirmed malicious by TPCI chrome-mal-ids — Sider: Chat with all AI: GPT-5, Claude, DeepSeek, Gemini, Grok [data-theft] + malext.io malicious_extension_sentry (+2.5, capped).
- Privacy policy fetch failed — no verifiable data-handling disclosure for 5M-user AI extension with <all_urls> access.
- cookies + <all_urls> + scripting combo: can read session cookies and inject JS on every site the user visits.
- 20 dom_sink_innerhtml_userctrl findings across content scripts expose DOM-XSS attack surface on all visited pages.
- brand_mention.is_impersonation=true: extension name drops Claude, Gemini, DeepSeek branding without confirmed ownership.
Evidence
- host_permissions_broad manifest <all_urls> + cookies + scripting = full page read/write and cross-site cookie access on every URL.
- privacy_policy_fetch_failed api privacy_policy_classification.fetched==false (HTTPError); policy content unverifiable → pillar scores 10.0.
- brand_impersonation store brand_mention.is_impersonation=true; brands: deepseek, claude, gemini; confirmed_owner=false; not verified publisher.
- uninstall_url_hijack crx uninstall_url_hijack=true; redirects to unspecified 3rd-party URL on uninstall.
- dom_xss_sinks_widespread crx 16 files contain dom_sink_innerhtml_userctrl; CSP present but sandbox allows external GA/Facebook/Twitter JS.
- function_constructor crx new Function() found in 4 api.js files under js/v2/ — dynamic code execution risk.
- csp_sandbox_external_scripts manifest Sandbox CSP allows ssl.google-analytics.com, connect.facebook.net, platform.twitter.com — remote script sources.
- no_developer_name store developer_name is empty string; reduces accountability; email care@sider.ai on resolving domain sider.ai.
- threat_feed_membership api Extension ID listed by TPCI chrome-mal-ids — Sider: Chat with all AI: GPT-5, Claude, DeepSeek, Gemini, Grok [data-theft] + malext.io malicious_extension_sentry; score 4.50 -> 7.00 (+2.5 cap 2.5).
Permissions Breakdown
- storage low Stores extension state locally.
- cookies high Can read/write cookies across all sites given <all_urls> host access.
- contextMenus low Adds right-click menu items; low standalone risk.
- scripting high Programmatic script injection on any page with <all_urls>.
- activeTab medium Grants temporary access to current tab on user action.
- unlimitedStorage low No data-exfil capability, only local storage quota.
- tabs medium Reads URL/title of all open tabs.
- sidePanel low UI surface only.
- offscreen low Background DOM processing; limited standalone risk.
- alarms low Periodic wake-ups; no direct data access.
- declarativeNetRequest medium Can modify/block network requests declaratively.
- <all_urls> (host) high Broad host access amplifies cookies+scripting to every site visited.
- https://*.openai.com/ (host) high Explicit access to OpenAI sessions and API traffic.
Pillar Scores
Permissions7.80
Reputation5.50
Network4.00
Webstore5.50
Maintenance0.00
Privacy10.00
Code Quality5.00
CVE Exposure0.00
Bookkeeping
Rubric v3.6
Scored at 2026-11-03 09:52
Listing SHA
ab66f187ca10…
Force block
— not fired
Score recovered
no
Elapsed
34.4s