obsidian-migration-deep-dive
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external note exports (e.g., ENEX, JSON, CSV). While this represents an attack surface for indirect prompt injection, it is an essential component of the migration functionality and includes basic sanitization for file paths.\n
- Ingestion points: Reads exported content from multiple note platforms in various formats including Markdown, HTML, XML, and JSON.\n
- Boundary markers: Does not implement specific delimiters to isolate untrusted data, which is typical for conversion scripts of this nature.\n
- Capability inventory: Utilizes file system read/write and shell execution for data processing.\n
- Sanitization: Employs filename sanitization to ensure compatibility with local file systems.\n- [DYNAMIC_CONTEXT_INJECTION]: The
SKILL.mdfile contains!commandmarkers for environment assessment upon loading. These markers perform benign checks such as verifying the Node.js version and listing potential source files in the working directory.\n- [DYNAMIC_EXECUTION]: The migration process involves generating and executing local Node.js and Bash scripts (e.g.,notion-to-obsidian.mjs). This dynamic execution is standard for providing customizable migration logic within the skill environment.\n- [COMMAND_EXECUTION]: The skill includes shell commands for environment assessment, dependency installation vianpm, and running conversion utilities. This includes an inline Node.js script for converting Apple Notes HTML to Markdown.
Audit Metadata