vault-qmd

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill facilitates the execution of local qmd commands (status, query, search, get, ls, update, embed) for vault management.
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface: 1. Ingestion points: Reads vault notes via qmd query, qmd search, and qmd get (SKILL.md). 2. Boundary markers: No explicit use of delimiters or ignore instructions for note content. 3. Capability inventory: Ability to update indices and append to logs via qmd update, vault-index, and vault-log (SKILL.md). 4. Sanitization: No mentioned validation or sanitization of retrieved markdown content.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 12:47 AM
Security Audit — agent-trust-hub — vault-qmd