skills/joelgwebber/yaks/yak-tracker/Gen Agent Trust Hub

yak-tracker

Pass

Audited by Gen Agent Trust Hub on Jul 26, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection (Category 8) by ingesting data from external trackers that could contain malicious instructions.\n
  • Ingestion points: Fetches titles, descriptions, and comments from external issues (SKILL.md).\n
  • Boundary markers: Absent; the skill does not instruct the agent to delimit or ignore instructions within the fetched content.\n
  • Capability inventory: Includes ability to create/update local yaks and write to external trackers via MCP tools and the GitHub CLI.\n
  • Sanitization: Absent; no mention of filtering or escaping external content before interpolation.\n- [COMMAND_EXECUTION]: The skill relies on local and external command-line tools for core functionality.\n
  • Evidence: Uses the yaks CLI for data management and the gh CLI for interacting with GitHub repositories.\n- [EXTERNAL_DOWNLOADS]: The skill fetches data from well-known services to sync issue status.\n
  • Evidence: References Atlassian (Jira), Linear, and GitHub APIs to retrieve issue metadata and comments.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 26, 2026, 12:35 AM
Security Audit — agent-trust-hub — yak-tracker