x9-agent-instructions
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script
check_globals.pyto verify that specific blocks in configuration files (~/.config/opencode/AGENTS.md,~/.claude/CLAUDE.md, and~/.codex/AGENTS.md) are identical. This is a local integrity check relevant to the skill's purpose of managing agent instructions. - [DATA_EXPOSURE]: The script accesses standard configuration paths for AI agents (Claude Code, OpenCode, Codex) located in the user's home directory. This access is restricted to reading and comparing specific markers (
<!-- BEGIN SHARED PERSONAL CORE -->) and does not involve network transmission or credential harvesting.
Audit Metadata