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 Apr 7, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill is designed to access and process sensitive communication data within Outlook PST and OST files, including emails, contacts, and metadata. This data access is localized to the user's environment, involving reading from known Outlook paths and writing results to a local output directory for investigative purposes.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from analyzed emails, creating an indirect prompt injection surface.
- Ingestion points: The PSTForensicAnalyzer class in SKILL.md and the parsing logic in scripts/agent.py ingest email subjects, bodies, and transport headers from external PST/OST files.
- Boundary markers: Absent. No explicit delimiters or instructions are provided to the agent to disregard potentially malicious content embedded within the ingested email data.
- Capability inventory: The skill utilizes file system access to read PST files and write extracted attachments and reports as evidenced in SKILL.md and scripts/agent.py.
- Sanitization: Includes basic alphanumeric filtering for attachment filenames in SKILL.md to prevent filesystem-related issues, but does not sanitize email content for injection vectors.
Audit Metadata