red-handed-audit

Fail

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the @jinhyuk9714/red-handed package from the public NPM registry at runtime using npx --yes. This package comes from an unverified third-party source.
  • [REMOTE_CODE_EXECUTION]: The skill executes the downloaded package locally via npx. This allows unverified remote code to run with the permissions of the local user and access sensitive session data.
  • [COMMAND_EXECUTION]: The skill invokes the shell to execute npx commands. These commands are configured to read the Claude Code session transcript and the project's git working tree, which are sensitive local data sources.
  • [DATA_EXFILTRATION]: While the skill claims no data leaves the machine, the execution of unverified third-party code that reads sensitive local transcripts creates a risk of data exfiltration if the package or its dependencies are compromised.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Aug 11, 2026, 01:47 PM
Security Audit — agent-trust-hub — red-handed-audit