obsidian-workflows

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The file references/advanced-workflows.md contains a shell script snippet for vault maintenance that uses standard Unix commands such as find, wc, sort, and uniq to analyze vault file metadata.
  • [PROMPT_INJECTION]: The skill defines workflows for Inbox Processing and Weekly Review that involve an agent reading and making decisions based on the content of user-provided notes, which presents a surface for indirect prompt injection.
  • Ingestion points: Content from user notes located in the 0
  • Inbox and PARA folders (defined in SKILL.md).
  • Boundary markers: None identified; instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing note content.
  • Capability inventory: The agent is authorized to modify vault structure, move files, and edit note contents during review processes.
  • Sanitization: There is no mention of sanitizing or validating note content before processing or acting upon it.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 07:03 AM