update

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • Command Execution: The skill utilizes the GitHub CLI (gh issue list --assignee=@me) to fetch tasks assigned to the current user. This is a standard integration for task management.
  • External Data Retrieval: In its comprehensive mode, the skill scans communication platforms like chat and email via the Model Context Protocol (MCP). This process retrieves data from external services to enrich the local memory.
  • Indirect Prompt Injection Surface: The skill processes untrusted content from external sources which could potentially contain instructions. The skill handles this through an interactive workflow requiring user confirmation for all changes.
  • Ingestion points: TASKS.md, GitHub issues, and communication platforms (via MCP).
  • Boundary markers: Employs interactive user confirmation rather than explicit data sandboxing.
  • Capability inventory: Reading/writing local markdown files and executing the gh command.
  • Sanitization: No explicit sanitization of external content is specified before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 11:59 PM