memorb-domain-query

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions specify the use of the grep command-line utility to perform searches across local directories within the user's vault.
  • Evidence: grep -ri "domain:.*acme.com" "$VAULT/memorbs/organizations/" --include="*.md" and grep -ri "acme.com" "$VAULT/memorbs/people/" --include="*.md" in SKILL.md.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it retrieves and processes data from local markdown files to drive downstream actions, such as constructing queries for Outlook/M365.
  • Ingestion points: Markdown files located in memorbs/organizations/ and memorbs/people/ (referenced in SKILL.md).
  • Boundary markers: Absent. The instructions do not define specific delimiters or directives to ignore instructions that might be embedded within the retrieved vault notes.
  • Capability inventory: The skill utilizes shell command execution (grep), Dataview query execution, and external Outlook/M365 search capabilities.
  • Sanitization: No explicit sanitization, escaping, or validation of the data extracted from the vault notes is described before the data is used to perform searches or summarize emails.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 09:48 AM
Security Audit — agent-trust-hub — memorb-domain-query