skills.rest
Miasin-Labs
Miasin-LabsOFFICIAL
Explore all skills by this author →
GitHub

dream

Consolidates daily session logs and memories into organized topic files and learnings.

9|2|Updated May 3, 2026
Search Tags:#learnings#agent memory#scheduled task#memory consolidation#knowledge pruning#session transcripts#nightly reflection
One-click install
npx skills add https://github.com/Miasin-Labs/jfc --skill dream-miasin-labs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dream
Source: https://github.com/Miasin-Labs/jfc/tree/main/crates/jfc-agents/builtin-skills/claude-2.1.167/frontmatter-skills/dream
Command: npx skills add https://github.com/Miasin-Labs/jfc --skill dream-miasin-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Long-running AI assistants accumulate scattered session transcripts and daily logs that are lost when context resets. This Skill performs nightly reflection to consolidate that raw history into durable, organized memory files. ## Core Features & Use Cases - Memory Consolidation: Reviews daily logs and session transcripts, then extracts significant events, decisions, and insights into topic files. - Learning Capture: Records painful or inefficient experiences (failed builds, user frustration) into dedicated learnings files to avoid repeating mistakes. - Memory Pruning: Keeps the main MEMORY.md index under 200 lines by prioritizing, summarizing, and removing stale entries. - Use Case: An agent runs this as a scheduled overnight task (1-5am) so it wakes up with a curated, contradiction-free memory base instead of raw transcripts. ## Quick Start Schedule the dream skill to run nightly so it reviews today's logs and updates my memory files with new topics and learnings.

Frequently Asked Questions about dream

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

FAQPage Schema
How do I give an AI agent long-term memory across sessions?▼

Use a nightly consolidation routine that reviews daily logs and session transcripts, then writes durable topic files and learnings to a memory directory. This Skill automates that reflection process so knowledge survives context resets.

How to schedule an AI agent housekeeping task overnight?▼

This Skill is designed to run via a scheduled task scaffold between 1-5am local time. It operates as a forked context job and does not message the user unless it finds something noteworthy.

What file structure does the agent memory system use?▼

Memory lives under a MEMORY_ROOT path with dated logs in logs/YYYY/MM/, session transcripts in sessions/YYYY/MM/, topic files as top-level markdown, learnings in a learnings/ subdirectory, and a MEMORY.md index capped at 200 lines.

How does the skill avoid duplicate or contradictory memories?▼

During Phase 1 it reviews existing topics before writing, improving files that already cover a subject rather than creating duplicates. Phase 2 explicitly resolves contradictions found across the day's records.

What are the limitations of file-based agent memory?▼

File-based memory depends on disciplined pruning to stay useful, which is why MEMORY.md is capped at 200 lines. It also requires the scheduled task to actually run; skipped nights leave gaps in the consolidated record.

Related Skills

Explore More Software Engineering Skills→
qa-amanqa-aman
11

work-log

Generate daily work logs from Claude Code JSONL session transcripts.

Community
Intermediate
cookyscookys
10

learn

Record reusable lessons and fixes into MEMORY.md for future sessions.

Community
Intermediate
blackplume233blackplume233
12

lifecycle-policy

Define and enforce retention, backup, expiration, and cleanup policies for memories and assets.

Community
Intermediate
HeadyAIHeadyAI
1

heady-episodic-memory

Organize and retrieve multi-tier AI agent memories with phi-decay forgetting.

Official
Advanced