skills/elvisun/newsjack/pr-calendar/Gen Agent Trust Hub

pr-calendar

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to construct and execute shell commands (e.g., newsjack news search) using variable data such as event names and dates retrieved from external APIs. This creates a surface for command injection if the underlying tool execution does not properly escape shell characters.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from external sources (Medialyst PR calendar API and news search results).\n
  • Ingestion points: Medialyst PR calendar API response via newsjack pr-calendar query or MCP tool; News search results via newsjack news search (SKILL.md).\n
  • Boundary markers: None specified in the instructions for separating external data from agent instructions.\n
  • Capability inventory: Subprocess execution via newsjack CLI and MCP network operations (SKILL.md).\n
  • Sanitization: No explicit sanitization or filtering of external content before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 01:32 AM
Security Audit — agent-trust-hub — pr-calendar