compound-engineering

Scan agent sessions and consolidate learnings into MEMORY.md files.

Updated Feb 7, 2026
One-click install
npx skills add https://github.com/tarotmansa/moltguild --skill compound-engineering-tarotmansa
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: compound-engineering
Source: https://github.com/tarotmansa/moltguild/tree/main/skills/compound-engineering
Command: npx skills add https://github.com/tarotmansa/moltguild --skill compound-engineering-tarotmansa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Removes the manual, ad-hoc process of reviewing agent sessions and maintaining long-term memory by automating nightly and hourly review loops that extract learnings and update memory files so the agent improves over time.

Core Features & Use Cases

  • Automated Nightly Review: Scan all sessions from the last 24 hours, extract key learnings, and update long-term and daily memory files.
  • Hourly Snapshots: Create frequent memory snapshots to capture incremental progress and decisions.
  • Extracted Artifacts: Identify patterns, gotchas, user preferences, key decisions, and open items for follow-up.
  • Integrations: Designed to run via cron, Clawdbot, or launchd and commit memory updates to git for versioned agent instructions.
  • Use Case: Run a nightly compound job to consolidate a day's sessions into MEMORY.md and memory/YYYY-MM-DD.md so tomorrow's agent benefits from yesterday's insights.

Quick Start

Run the compound-engineering review command to scan the last 24 hours, extract key learnings, and update MEMORY.md and today's memory file.

Frequently Asked Questions about compound-engineering

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate nightly agent session reviews to update long-term memory?

Automate nightly agent session reviews by scanning the last 24 hours of sessions, extracting patterns, and updating long-term memory files like MEMORY.md and daily memory logs. This consolidation ensures tomorrow's agent benefits from yesterday's insights.

What is the best way to extract agent learnings and commit them to a git repository?

Extract agent learnings and commit them to a git repository by scanning recent sessions for patterns, gotchas, and decisions, then writing these extracted artifacts to structured memory files. The automated process versions agent instructions for continuous improvement.

Can I run hourly memory snapshots to capture incremental agent progress?

Yes, you can run hourly memory snapshots to capture incremental agent progress and decisions. Frequent snapshots create structured memory files that record ongoing progress, ensuring no intermediate learning is lost between nightly reviews.

Does this automated memory consolidation work with cron, launchd, or Clawdbot jobs?

Automated memory consolidation is designed to work with cron, launchd, and Clawdbot jobs. You can schedule nightly reviews or hourly snapshots through these schedulers to automatically scan sessions and commit memory updates.

What specific artifacts are extracted during an automated agent memory review?

Automated agent memory reviews extract specific artifacts like patterns, gotchas, user preferences, key decisions, and open items for follow-up. These extracted learnings are written to structured memory files for future agent reference.

How do I structure memory files to help an AI agent improve over time?

Structure memory files by updating a central MEMORY.md for long-term learnings and creating daily memory/YYYY-MM-DD.md files for session-specific insights. This dual-layer structure allows agents to compound knowledge and improve over time.