ljg-writes
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The instructions contain stylistic directives and constraints, such as 'Internalize, don't quote,' which are standard prompt engineering techniques for maintaining output consistency and confidentiality of local reference files. These do not represent malicious attempts to bypass AI safety filters.
- [DATA_EXFILTRATION]: The skill interacts with the local file system at '
/Documents/know/soul.md' (read) and '/Documents/notes/' (write). These actions are aligned with the skill's primary function as a writing and note-taking tool. No external network requests or sensitive credential access were detected. - [COMMAND_EXECUTION]: The skill references the 'date' utility's formatting syntax to define identifiers and timestamps for file naming. These are placeholders for generating dynamic text and do not involve the execution of arbitrary or dangerous system commands.
Audit Metadata