vault-synthesize

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it reads and synthesizes information from untrusted local vault files (such as README.md, gemini3_summary.txt, and transcript.txt). \n
  • Ingestion points: Data enters the agent context through the Read tool in Step 4. \n
  • Boundary markers: The synthesis template in Step 5 does not utilize specific delimiters or instructions to ignore instructions embedded within the source documents. \n
  • Capability inventory: The skill is equipped with Read and Write tools, alongside the obsidian CLI for searching and logging to daily notes. \n
  • Sanitization: There is no evidence of sanitization or validation of the content retrieved from the vault files prior to its use in generating new notes.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 09:09 AM