index

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements a workflow for indexing external data, which presents a surface for indirect prompt injection if the indexed files contain instructions that influence the agent's behavior during search retrieval.
  • Ingestion points: Files from the user's local vault are ingested and processed by the obsidian_rag_index tool described in SKILL.md.
  • Boundary markers: No delimiters or protective instructions are provided to help the agent distinguish between data and embedded instructions within the vault files.
  • Capability inventory: The skill is configured to call the obsidian_rag_index tool, which accesses and processes file content.
  • Sanitization: The instructions do not specify any sanitization, filtering, or validation of the file content before it is added to the semantic index.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 10:47 AM
Security Audit — agent-trust-hub — index