SECURITY REPORT · EXTENSHI SCAN
Verbatim: Adversarial Review for AI
Chrome extension·by Verbatim
0
/ 100 SAFETY SCORE
RISKY
Several elevated signals. Read the findings before sharing this with your team.
Automated analysis of publicly available code — not a professional security audit. It may produce false positives and does not prove an extension is unsafe or safe. Treat it as one signal among many. Disclaimer · Report an issue
How it ranks
Verbatim: Adversarial Review for AI's safety score vs every other extension we've scanned
Globally
5%of extensions carry more risk
Riskier ←→ Safer
Ranked against 120,712 extensions
In Productivity
12%of productivity extensions carry more risk
Riskier ←→ Safer
Ranked against 25,973 productivity extensions
Findings
117
median 10Lower is safer
Critical findings
20
median 0Severity = Critical
Permissions
2
median 3Declared in manifest
Active users
4
median 23vs category median
Findings breakdown
Automated security signal · 117 findings across 5 severities
117
findings
detected
detected
Critical2017%
High11%
Medium9682%
Low00%
Info00%
Permissions requested
2 permissions declared
No risky permissions
2 permissions requested — none are flagged as risky.
Medium
Store local data
Can save data locally in the browser for persistence
Medium
Access user identity
Can get OAuth2 tokens and access your Google account identity
Are you the author of Verbatim: Adversarial Review for AI?
Claim this extension to respond to findings, add context, and earn a verified badge on extenshi.io.
Scan performed by extenshi.io · Jun 25, 2026
Browser extensions made transparent.