wiki-graph-query

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a structured protocol for exploring a knowledge graph using an internal tool. No security risks were identified.
  • [COMMAND_EXECUTION]: The skill utilizes the vault_query_graph tool. This access is constrained by a 'HARD-GATE' requiring the agent to verify that the target node exists as a file within the vault before execution, reducing the risk of invalid tool calls.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: There are no network calls or patterns suggesting data exfiltration. The traversal is limited to the internal knowledge graph environment.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes external node paths, it includes a validation step ('Confirm node file exists in vault') to ensure only existing vault content is used as a traversal seed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 07:04 AM