token-dashboard-claude-analytics
Fail
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires cloning a repository from an unverified GitHub account (nateherkai/token-dashboard) that does not match the stated author patterns for the vendor.
- [REMOTE_CODE_EXECUTION]: Instructs users to execute Python scripts (python3 cli.py) from the downloaded repository immediately, which can result in arbitrary code execution without verification.
- [DATA_EXFILTRATION]: The skill is designed to read and process highly sensitive files containing private chat transcripts and session history located in ~/.claude/projects/. While claimed to be local-only, the code is unverified.
- [COMMAND_EXECUTION]: Uses system commands to scan local directories and host a web server, which could be misconfigured to expose sensitive transcript data to the local network.
Recommendations
- AI detected serious security threats
Audit Metadata