Security context
What an agent needs to avoid regressing past fixes and find the next vuln in this repo.
jank-lang/jank
main @ 1555bc6
3
Fixes
1
CVEs
HIGH
Peak severity
Highlights
Arbitrary Code Execution: 1 prior fix. Scrutinize any change in this area.
.github/workflows/build.yml: 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.
Arbitrary Code Execution: The runtime context allowed unsafe Clojure reader evaluation during string and file parsing operations. Attackers capable of feeding untrusted input to the reader could trigger remote code execution if context-level variables like *read-eval* are not strictly disabled.
Privilege Escalation: Overly permissive GITHUB_TOKEN configurations on CI/CD workflows expose the repository to privilege escalation if a runner or dependent action is compromised. Enforcing 'contents: read' minimizes the target surface.
Information Disclosure: Temporary private SSH keys used during builds were written with overly broad permissions, exposing sensitive credentials to other processes or users sharing the execution context.