Back to skill
Skillv1.1.1
VirusTotal security
Improvement Evaluator · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignApr 7, 2026, 3:31 AM
- Hash
- 8cc18bf1c1ae97437759d4469aab5b5ac549fb058f79c2c93806158b28d08b2f
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: improvement-evaluator Version: 1.1.1 The bundle implements an evaluation framework for measuring the effectiveness of AI skill improvements. It uses subprocess execution to run 'pytest' for structured output validation and the 'claude' CLI for rubric-based scoring; while these are high-privilege operations, they are functionally necessary for the stated purpose. The implementation in 'interfaces/judges.py' includes explicit security mitigations, such as path traversal guards and symlink resolution, to ensure that 'pytest' only executes code within the designated 'fixtures/' directory, demonstrating a defensive posture rather than malicious intent.
- External report
- View on VirusTotal