consult-status-report

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources which could contain malicious instructions designed to influence the agent's summary or the generated HTML.
  • Ingestion points: CRM deal notes and activity logs (SKILL.md Step 2); client documents including [Notes], [Plan], [BRD], and [UAT] files from Google Drive (SKILL.md Step 3).
  • Boundary markers: The skill lack explicit delimiters for prompt interpolation of external data, though it instructions the agent to "skim for dated facts" rather than full ingestion.
  • Capability inventory: The skill can create Gmail drafts (mcp__google-workspace__draft_gmail_message) and update CRM records (mcp__zynkr__update_deal).
  • Sanitization: The accompanying Python script (scripts/render_dashboard_email.py) utilizes html.escape for all data fields before rendering them into the final HTML template, which provides defense against XSS and structural manipulation in the email body.
  • [COMMAND_EXECUTION]: The workflow involves executing a local Python script to transform the generated JSON report into an HTML email body.
  • Evidence: python3 ./scripts/render_dashboard_email.py dashboard.json --out email.html (SKILL.md Step 7).
  • [EXTERNAL_DOWNLOADS]: The skill documentation includes a command to fetch components from a remote repository owned by the author.
  • Evidence: npx skills add https://github.com/peter-tu-zynkr/zynkr-skill-builder --skill consult-status-report (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 07:30 PM
Security Audit — agent-trust-hub — consult-status-report