evernote-migration-deep-dive

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The SKILL.md file contains a dynamic context injection pattern (!npm list 2>/dev/null | head -5) that executes a shell command when the skill is loaded. This is used for benign environment introspection and is consistent with the skill's developer-focused migration purpose.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it handles external data from Evernote.
  • Ingestion points: Data enters the system context through the Evernote API (notes, notebooks, tags) and exported local files as described in references/implementation-guide.md.
  • Boundary markers: There are no explicit delimiters or boundary markers used in the code to isolate external note content from agent instructions.
  • Capability inventory: The skill is configured with Write, Edit, and Bash tools and includes scripts that execute filesystem operations (fs.writeFile, fs.mkdir).
  • Sanitization: The implementation includes the sanitizeName method for output filenames in the EvernoteExporter class, but no instructional sanitization is applied to the note content itself.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 05:16 AM