validator-run
Warn
Audited by Socket on May 17, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill's purpose and local file operations are broadly coherent for a validator runner, and there is no explicit credential theft or off-platform exfiltration. However, the central executable `agent-validate` is not verifiably sourced or documented from the available evidence, and the skill combines autonomous Bash execution with log-driven fix loops, which raises meaningful supply-chain and prompt-injection risk.
Confidence: 87%Severity: 72%
Audit Metadata