vault-trace

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill reads and summarizes content from various notes in the user's vault, which creates a surface for indirect prompt injection. If notes contain malicious instructions, they could influence the agent's summary or sentiment analysis.
  • Ingestion points: Content is ingested from file bodies, filenames, tags, and wikilinks across the vault as specified in Step 1.
  • Boundary markers: No explicit delimiters are used to separate the note content from the agent's instructions.
  • Capability inventory: The skill can search and read files using glob/grep and write synthesis pages to the 'notes/concepts/' directory. No network or administrative capabilities are present.
  • Sanitization: No sanitization or filtering of the extracted note content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 03:46 PM
Security Audit — agent-trust-hub — vault-trace