sdlc
Warn
Audited by Socket on Sep 21, 2026
1 alert found:
AnomalyAnomalytemplates/ci/sdlc-evals.yml
LOWAnomalyLOW
templates/ci/sdlc-evals.yml
The workflow is a legitimate CI evaluation job and contains no direct evidence of malware. Its main security concern is supply-chain exposure from installing the latest unpinned @anthropic-ai/claude-code package and then running it with an ANTHROPIC_API_KEY while processing repository-controlled configuration. The code fragment alone does not demonstrate credential theft or malicious execution, but the design creates a meaningful dependency and secret-exposure risk.
Confidence: 97%Severity: 62%
Audit Metadata