normalize-format
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data sources, including Claude.ai exports, Readwise highlights, and transcripts. While this represents a surface for indirect prompt injection, the skill's logic is strictly limited to text formatting and structural normalization. It does not contain instructions to execute commands, perform network operations, or access sensitive files based on the content of the processed data.
- [SAFE]: The skill consists entirely of instructional text without any embedded scripts or executable patterns. No obfuscation, data exfiltration attempts, or privilege escalation patterns were detected. The operations described (parsing, splitting text, and stripping timestamps) are benign and consistent with the skill's stated purpose.
Audit Metadata