cold-email
Pass
Audited by Gen Agent Trust Hub on May 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected in the skill's instructions or scripts. All external links in the deliverability guide point to well-known and reputable services for email testing and domain reputation.
- [COMMAND_EXECUTION]: The skill includes a Python utility (
scripts/email_sequence_analyzer.py) for evaluating email quality. Analysis of the source code confirms it uses only standard libraries, does not perform network operations, and limits its file system access to reading the provided input file. - [PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection because it is designed to process user-supplied email drafts and performance data. However, the risk is negligible as the skill lacks capabilities for network exfiltration or sensitive system modification, and its primary function is benign text analysis.
- Ingestion points: Mode 1 (Write), Mode 2 (Build), and Mode 3 (Iterate) ingest user-supplied text; the Python script ingests JSON via file or stdin.
- Boundary markers: Not explicitly defined in the logic.
- Capability inventory: Python execution for text analysis; no network or file-write capabilities.
- Sanitization: None performed on the input text.
Audit Metadata