skills/mukul975/anthropic-cybersecurity-skills/analyzing-outlook-pst-for-email-forensics/Gen Agent Trust Hub
analyzing-outlook-pst-for-email-forensics
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a legitimate forensic toolset with no evidence of malicious intent. All file operations are local and consistent with the stated purpose.
- [EXTERNAL_DOWNLOADS]: The documentation instructs the user to install well-known forensic libraries including 'libpff-python' and 'libratom' using standard package managers.
- [COMMAND_EXECUTION]: The skill includes usage examples for 'pffexport' and 'pffinfo', which are standard open-source command-line utilities for PST analysis.
- [DATA_EXFILTRATION]: The provided Python scripts perform local analysis and do not contain any network-related code or exfiltration patterns.
- [PROMPT_INJECTION]: The skill handles untrusted PST data, posing a surface for indirect prompt injection. Ingestion points: PST files processed in 'agent.py' and 'SKILL.md'. Boundary markers: No explicit boundary markers between data and agent instructions. Capability inventory: Local file system write for attachments and forensic reports. Sanitization: Filename cleaning is implemented to prevent path traversal during attachment extraction.
Audit Metadata