obsidian-dashboard

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill suggests including Chart.js from the jsDelivr CDN (https://cdn.jsdelivr.net/npm/chart.js) to provide visualizations within the dashboards. This is a well-known and reputable service for hosting open-source libraries.
  • [COMMAND_EXECUTION]: Instructions mention the 'clawg dashboard' command to serve dashboards locally. This is a project-specific tool for visualization and does not involve suspicious privilege escalation.
  • [PROMPT_INJECTION]: The skill is designed to read existing HTML files from the local vault to perform updates, which creates an indirect prompt injection surface where untrusted data could influence agent behavior. 1. Ingestion points: The agent reads existing dashboard HTML files from the 'dashboard/' directory in the local vault. 2. Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands during the file reading process. 3. Capability inventory: The skill has file-write access to the vault and mentions a local server command. 4. Sanitization: No explicit sanitization or validation of the content read from the files is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 03:12 PM