Security context
What an agent needs to avoid regressing past fixes and find the next vuln in this repo.
tisfeng/easydict
dev @ 2d7c999
5
Fixes
0
CVEs
HIGH
Peak severity
Highlights
Auth Bypass: 1 prior fix. Scrutinize any change in this area.
ClaudeCodeRunner: most-fixed (1 issue). Treat as high-risk during review.
1 high-severity fix in this history; regressions here are high-impact.
Recurring patterns
The bug types that recur here, drawn from past fixes, not open vulnerabilities.
Auth Bypass: A vulnerability in the Sparkle framework could allow signature verification bypass, potentially enabling malicious updates to execute on the host machine.
Command Injection: Using unvalidated shell executables based on the environment's SHELL variable could allow arbitrary binary execution on the user system.
Sandbox Escape: Optional LLM agent tools like shell and browser access during translation tasks could allow models to break sandbox constraints via tool use.