LaunchKit · 2026
Back to Skills

super-proactive

**The ultimate proactive + memory system for AI agents.** Combines the best of 11 top-rated skills into one unified.

3
695 downloads
by @heldinhow

Setup & Installation

openclaw skills install @heldinhow/super-proactive

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install super-proactive

Version History

v1.0.1Feb 15, 2026 - Version note by skill developer:

- Documentation in SKILL.md was streamlined for clarity and consistency. - File structure section reformatted for improved readability. - Example commands, task samples, and file descriptions now use more generic, simplified language. - Some usage and best practice details were condensed; outdated ratings and verbose sections were removed. - No functional or code changes—documentation update only.

What This Skill Does

Combines memory architecture, task queuing, and autonomous scheduling into a single agent skill. Uses a three-tier memory system (episodic, semantic, procedural) alongside a WAL protocol that logs decisions before responding. Background crons run on 30-minute, 4-hour, and daily schedules without user prompting.

Consolidates 11 separate memory and proactivity skills into one setup, avoiding the need to configure and maintain them independently.

When to use it

  • Maintaining context across long multi-session projects
  • Tracking a backlog of research tasks autonomously
  • Logging decisions made during a debugging session
  • Running periodic health checks on a service without reminders
  • Preserving key learnings across context window resets

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: 'Pick up where we left off on the API integration project'

AGENT
  1. 1Read SESSION-STATE.md to restore current project context
  2. 2Search MEMORY.md for stored learnings about the API integration
  3. 3Check QUEUE.md for any Ready or In Progress tasks
  4. 4Resume the highest-priority In Progress task
  5. 5Log current state back to SESSION-STATE.md before responding
OUTPUT

Agent resumes the correct task with full context, no re-explanation needed