utilitry

Whitelist-based permission system for Claude Code skills. Declare what a skill is allowed to do in its SKILL.md frontmatter; anything not whitelisted falls through to the normal Claude Code permission flow. Includes a dangerous-command safety net and desktop notifications when human input is needed.

Capture and act on todos, reminders, and notes. Storage at ~/.notably/ as markdown files with YAML frontmatter. Reminders fire via SessionStart surfacing, scheduled remote agents, or macOS notifications. Slash-only invocation, no model auto-triggering.

Capture, browse, and consume agentic blindspots - default tendencies that produce bad outputs in specific contexts. Records live as markdown files with YAML frontmatter under ~/.blindspots/.