Security context

What an agent needs to avoid regressing past fixes and find the next vuln in this repo.

agentscope-ai/agentscope

              main @ ae98190
            

Fixes

CVEs

CRITICAL

Peak severity

0.0%

Coverage

Highlights

Auth Bypass: 2 prior fixes. Scrutinize any change in this area.

src/agentscope/app/_tools/_agent_create.py: most-fixed (1 issue). Treat as high-risk during review.

2 high-severity fixes in this history; regressions here are high-impact.

Recurring patterns

The bug types that recur here, drawn from past fixes, not open vulnerabilities.

Path Traversal: Insecure file path validation and weak blocklists for dangerous files/directories can allow malicious agents or inputs to write, edit, or execute files outside of the intended workspace. Strict path canonicalization and predefined safety boundaries must be enforced.

Auth Bypass: Flaws in the permission verification engine can allow unauthorized execution of dangerous tools. The validation logic must reliably enforce bypass-immune policy checks across all execution modes.

Auth Bypass: Spawning new worker agents without inheriting the parent's permission context, directory restrictions, and security rules allows newly created agents to run with unconstrained privileges.