← Back to Skills Marketplace
rimaslogic

Autodream

by Rimas Logic · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
153
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install autodream
Description
Automatic memory consolidation for OpenClaw agents. Analyzes daily memory files, removes duplicates, prunes stale entries, normalizes dates, and builds a cle...
README (SKILL.md)

Autodream — Memory Consolidation

Consolidates scattered daily memory files into a clean, organized MEMORY.md.

When to Use

  • Manually: User says "consolidate memory", "dream", "clean up memory", "organize my memories"
  • Via heartbeat: Periodically check if consolidation is needed (24h+ since last run AND 5+ new files)
  • After major events: Big refactor, project changes, many new daily files

Quick Start

# Check if consolidation is needed
node /path/to/openclaw-autodream/bin/autodream.js {{workspace}} --stats

# Run consolidation (dry run first!)
node /path/to/openclaw-autodream/bin/autodream.js {{workspace}} --dry-run --verbose

# Run for real
node /path/to/openclaw-autodream/bin/autodream.js {{workspace}} --verbose

# Force full reconsolidation
node /path/to/openclaw-autodream/bin/autodream.js {{workspace}} --force --verbose

Replace {{workspace}} with your actual workspace path (e.g., ~/.openclaw/workspace).

If Installed Globally

# npm install -g openclaw-autodream
autodream ~/.openclaw/workspace --stats
autodream ~/.openclaw/workspace --verbose
autodream ~/.openclaw/workspace --dry-run

What It Does (4 Phases)

Phase 1: Orientation

  • Reads existing MEMORY.md (if any)
  • Scans memory/ for daily files
  • Identifies files changed since last consolidation

Phase 2: Gather Signal

  • Extracts structured entries from each daily file
  • Classifies into categories: People, Projects, Preferences, Technical Decisions, Events, Lessons
  • Scores importance of each entry

Phase 3: Consolidation

  • Removes exact duplicates (hash-based)
  • Removes fuzzy duplicates (similarity > 75%)
  • Prunes stale entries (completed tasks, old debugging notes)
  • Normalizes relative dates ("yesterday" → "2026-03-24")

Phase 4: Prune & Index

  • Enforces max line limit (default: 200 lines)
  • Prioritizes by importance × recency
  • Writes clean MEMORY.md with category sections
  • Backs up previous MEMORY.md to memory/.autodream-backups/

Heartbeat Integration

Add to your HEARTBEAT.md:

## Memory Consolidation Check
- Run `autodream \x3Cworkspace> --stats` to check if consolidation is needed
- If "Would trigger: ✅ yes", run `autodream \x3Cworkspace> --verbose`
- Only run consolidation during quiet hours (not while human is actively chatting)

Configuration

Create .autodream.json in workspace root to customize:

{
  "maxLines": 200,
  "lookbackDays": 30,
  "categories": [
    "People & Relationships",
    "Projects & Work",
    "Preferences & Style",
    "Technical Decisions",
    "Important Events",
    "Lessons Learned"
  ],
  "preservePatterns": ["⚠️", "IMPORTANT", "NEVER", "ALWAYS"],
  "triggerThreshold": {
    "minHoursSinceLastRun": 24,
    "minNewFiles": 5
  }
}

Safety

  • Non-destructive: Always backs up existing MEMORY.md before modifying
  • Backups: Stored in memory/.autodream-backups/
  • Reports: Consolidation reports in memory/.autodream-reports/
  • Dry run: Always available with --dry-run
  • Protected entries: Entries with ⚠️, IMPORTANT, NEVER, ALWAYS are never pruned

Output Format

# Long-Term Memory
\x3C!-- Last consolidated: 2026-03-25T17:00:00Z | Files processed: 31 | Entries: 47 -->

## People & Relationships
- **Bob Smith** — Team Lead test run passed (2026-03-25)

## Projects & Work
- **Acme Corp** — Q1 review: 30.1% margin, targeting 40% (2026-03-25)

## Preferences & Style
- Values precision and factual accuracy (2026-02-01)

## Technical Decisions
- Using Supabase + Vercel for Project Alpha (2026-03-18)

## Lessons Learned
- Always backup before modifying production data (2026-02-15)
Usage Guidance
This is an instruction-only helper that expects a Node-based CLI (openclaw-autodream) to be installed separately; the skill itself contains no code. Before running: (1) install and inspect the openclaw-autodream package source (npm or GitHub) so you know exactly what the CLI will do, (2) run the tool with --dry-run and review the consolidation report, (3) verify that backups are saved where you expect (memory/.autodream-backups/), and (4) avoid enabling any automated heartbeat run until you confirm the tool's behavior. The skill does not ask for secrets or contact external endpoints in its instructions, but the external CLI you install might — review that package before trusting automated runs.
Capability Analysis
Type: OpenClaw Skill Name: autodream Version: 1.0.0 The autodream skill is a memory management utility designed to consolidate and prune agent memory files into a structured index. The documentation (SKILL.md) describes a legitimate multi-phase process for organizing workspace data, including safety features like automatic backups, dry-run modes, and the preservation of critical entries marked with specific keywords (e.g., 'IMPORTANT'). No malicious instructions, data exfiltration attempts, or suspicious code patterns were found in the provided metadata or markdown instructions.
Capability Assessment
Purpose & Capability
Name/description (memory consolidation) match the instructions: scanning memory/, reading/writing MEMORY.md, backing up to memory/.autodream-backups/, pruning entries and normalizing dates. Requested tool (node) is appropriate for a Node CLI; nothing asked-for is unrelated to the stated purpose.
Instruction Scope
SKILL.md instructs the agent to run a Node CLI (e.g., node /path/to/openclaw-autodream/bin/autodream.js or the global autodream command). All file reads/writes are scoped to the workspace (MEMORY.md, memory/). Instructions do not request unrelated env vars or system paths. Note: the skill assumes an external npm package/binary exists; since this skill is instruction-only and includes no code, the referenced CLI must be installed separately by the user for the commands to work.
Install Mechanism
No install spec or code files are bundled (instruction-only). The README suggests an npm package (openclaw-autodream) and invoking node. This is low-risk but means the actual executable is external and must be acquired/install by the user; no hidden downloads or extraction are present in the skill bundle itself.
Credentials
The skill declares no required env vars or credentials. It does not instruct accessing external secrets or unrelated configuration. Requested access is proportional: read/write within the user's workspace files only.
Persistence & Privilege
always is false, and the skill does not request persistent elevated privileges or modify other skills or system-wide settings. It writes backups and reports under memory/ in the workspace—this is within its stated scope.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install autodream
  3. After installation, invoke the skill by name or use /autodream
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Slug autodream
Version 1.0.0
License MIT-0
All-time Installs 1
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Autodream?

Automatic memory consolidation for OpenClaw agents. Analyzes daily memory files, removes duplicates, prunes stale entries, normalizes dates, and builds a cle... It is an AI Agent Skill for Claude Code / OpenClaw, with 153 downloads so far.

How do I install Autodream?

Run "/install autodream" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Autodream free?

Yes, Autodream is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Autodream support?

Autodream is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Autodream?

It is built and maintained by Rimas Logic (@rimaslogic); the current version is v1.0.0.

💬 Comments