skills/mukul975/anthropic-cybersecurity-skills/detecting-spearphishing-with-email-gateway/Gen Agent Trust Hub
detecting-spearphishing-with-email-gateway
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a defensive utility designed for cybersecurity analysis. No malicious patterns, obfuscation, or unauthorized operations were detected across the provided scripts and documentation.\n- [EXTERNAL_DOWNLOADS]: The documentation references official API endpoints and configuration pages for well-known security services such as Microsoft Defender for Office 365, Proofpoint, Mimecast, and Barracuda. These references are used for defensive configuration and documentation purposes.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of email files and gateway logs.\n
- Ingestion points: The scripts/agent.py and scripts/process.py scripts read and analyze external email headers, body content, and log entries.\n
- Boundary markers: The scripts perform local string and regex analysis; extracted content is not formatted as instructions for an LLM within the code.\n
- Capability inventory: The provided tools are limited to file system read access for analysis and standard output/file writing for reporting; no network exfiltration or shell command execution is automated on the untrusted data.\n
- Sanitization: Input processing utilizes standard libraries with error-handling for character decoding.
Audit Metadata