summary-user-said
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Python scripts to automate the discovery, parsing, and summarization of local chat history files.
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
esflowPython package to manage its processing workflow. - [DATA_EXFILTRATION]: The skill is designed to read sensitive local application data, specifically chat histories stored in
~/.cursor/projects,~/.claude/projects, and~/.codex/sessions. A thorough review of the code confirms that the skill has no networking capabilities and does not attempt to transmit this data externally. - [SAFE]: The skill demonstrates security best practices, including the use of exclusive file creation modes (
os.O_EXCL) and restrictive file permissions (0o600) for the generated reports. Additionally, it implements a robust validation layer that checks the AI's summary against original message hashes and source IDs to prevent hallucinations or data corruption.
Audit Metadata