OpenSkillIndex← back to search
claude-code

Skills tiered-memory

EvoClaw Tiered Memory Architecture v2.2.0 - LLM-powered three-tier memory system with automatic daily note ingestion, structured metadata extraction, URL preservation, validation, and cloud-first sync.

install
source · Clone the upstream repo
git clone https://github.com/openclaw/skills
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/openclaw/skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/bowen31337/tiered-memory" ~/.claude/skills/clawdbot-skills-tiered-memory && rm -rf "$T"
manifest: skills/bowen31337/tiered-memory/SKILL.md
source content

Tiered Memory System v2.2.0

A mind that remembers everything is as useless as one that remembers nothing. The art is knowing what to keep. 🧠

EvoClaw-compatible three-tier memory system inspired by human cognition and PageIndex tree retrieval.

What's New in v2.2.0

🔄 Automatic Daily Note Ingestion

  • Consolidation (
    daily
    /
    monthly
    /
    full
    modes) now auto-runs 
    ingest-daily
  • Bridges 
    memory/YYYY-MM-DD.md
    files → tiered memory system
  • No more manual ingestion required — facts flow automatically
  • Fixes the "two disconnected data paths" problem

What's New in v2.1.0

🎯 Structured Metadata Extraction

  • Automatic extraction of URLs, shell commands, and file paths from facts
  • Preserved during distillation and consolidation
  • Searchable by URL fragment

✅ Memory Completeness Validation

  • Check daily notes for missing URLs, commands, and next steps
  • Proactive warnings for incomplete information
  • Actionable suggestions for improvement

🔍 Enhanced Search

  • Search facts by URL fragment
  • Get all stored URLs from warm memory
  • Metadata-aware fact storage

🛡️ URL Preservation

  • URLs explicitly preserved during LLM distillation
  • Fallback metadata extraction if LLM misses them
  • Command-line support for adding metadata manually

Architecture

┌─────────────────────────────────────────────────────┐
│              AGENT CONTEXT (~8-15KB)                │
│                                                     │
│  ┌──────────┐  ┌────────────────────────────────┐  │
│  │  Tree    │  │  Retrieved Memory Nodes         │  │
│  │  Index   │  │  (on-demand, 1-3KB)            │  │
│  │  (~2KB)  │  │                                │  │
│  │          │  │  Fetched per conversation      │  │
│  │  Always  │  │  based on tree reasoning       │  │
│  │  loaded  │  │                                │  │
│  └────┬─────┘  └────────────────────────────────┘  │
│       │                                             │
└───────┼─────────────────────────────────────────────┘
        │
        │ LLM-powered tree search
        │
┌───────▼─────────────────────────────────────────────┐
│              MEMORY TIERS                           │
│                                                     │
│  🔴 HOT (5KB)      🟡 WARM (50KB)     🟢 COLD (∞)  │
│                                                     │
│  Core memory       Scored facts      Full archive  │
│  - Identity        - 30-day         - Turso DB     │
│  - Owner profile   - Decaying       - Queryable    │
│  - Active context  - On-device      - 10-year      │
│  - Lessons (20 max)                                │
│                                                     │
│  Always in         Retrieved via     Retrieved via │
│  context           tree search       tree search   │
└─────────────────────────────────────────────────────┘

Design Principles

From Human Memory

  • Consolidation — Short-term → long-term happens during consolidation cycles
  • Relevance Decay — Unused memories fade; accessed memories strengthen
  • Strategic Forgetting — Not remembering everything is a feature
  • Hierarchical Organization — Navigate categories, not scan linearly

From PageIndex

  • Vectorless Retrieval — LLM reasoning instead of embedding similarity
  • Tree-Structured Index — O(log n) navigation, not O(n) scan
  • Explainable Results — Every retrieval traces a path through categories
  • Reasoning-Based Search — "Why relevant?" not "how similar?"

Cloud-First (EvoClaw)

  • Device is replaceable — Soul lives in cloud (Turso)
  • Critical sync — Hot + tree sync after every conversation
  • Disaster recovery — Full restore in <2 minutes
  • Multi-device — Same agent across phone/desktop/embedded

Memory Tiers

🔴 Hot Memory (5KB max)

Purpose: Core identity and active context, always in agent's context window.

Structure:

{
  "identity": {
    "agent_name": "Agent",
    "owner_name": "User",
    "owner_preferred_name": "User",
    "relationship_start": "2026-01-15",
    "trust_level": 0.95
  },
  "owner_profile": {
    "personality": "technical, direct communication",
    "family": ["Sarah (wife)", "Luna (daughter, 3yo)"],
    "topics_loved": ["AI architecture", "blockchain", "system design"],
    "topics_avoid": ["small talk about weather"],
    "timezone": "Australia/Sydney",
    "work_hours": "9am-6pm"
  },
  "active_context": {
    "projects": [
      {
        "name": "EvoClaw",
        "description": "Self-evolving agent framework",
        "status": "Active - BSC integration for hackathon"
      }
    ],
    "events": [
      {"text": "Hackathon deadline Feb 15", "timestamp": 1707350400}
    ],
    "tasks": [
      {"text": "Deploy to BSC testnet", "status": "pending", "timestamp": 1707350400}
    ]
  },
  "critical_lessons": [
    {
      "text": "Always test on testnet before mainnet",
      "category": "blockchain",
      "importance": 0.9,
      "timestamp": 1707350400
    }
  ]
}

Auto-pruning:

  • Lessons: Max 20, removes lowest-importance when full
  • Events: Keeps last 10 only
  • Tasks: Max 10 pending
  • Total size: Hard limit at 5KB, progressively prunes if exceeded

Generates: 

MEMORY.md
— auto-rebuilt from structured hot state

🟡 Warm Memory (50KB max, 30-day retention)

Purpose: Recent distilled facts with decay scoring.

Entry format:

{
  "id": "abc123def456",
  "text": "Decided to use zero go-ethereum deps for EvoClaw to keep binary small",
  "category": "projects/evoclaw/architecture",
  "importance": 0.8,
  "created_at": 1707350400,
  "access_count": 3,
  "score": 0.742,
  "tier": "warm"
}

Scoring:

score = importance × recency_decay(age) × reinforcement(access_count)

recency_decay(age) = exp(-age_days / 30)
reinforcement(access) = 1 + 0.1 × access_count

Tier classification:

  • score >= 0.7
    → Hot (promote to hot state)
  • score >= 0.3
    → Warm (keep)
  • score >= 0.05
    → Cold (archive)
  • score < 0.05
    → Frozen (delete after retention period)

Eviction triggers:

  1. Age > 30 days AND score < 0.3
  2. Total warm size > 50KB (evicts lowest-scored)
  3. Manual consolidation

🟢 Cold Memory (Unlimited, Turso)

Purpose: Long-term archive, queryable but never bulk-loaded.

Schema:

CREATE TABLE cold_memories (
  id TEXT PRIMARY KEY,
  agent_id TEXT NOT NULL,
  text TEXT NOT NULL,
  category TEXT NOT NULL,
  importance REAL DEFAULT 0.5,
  created_at INTEGER NOT NULL,
  access_count INTEGER DEFAULT 0
);

CREATE TABLE critical_state (
  agent_id TEXT PRIMARY KEY,
  data TEXT NOT NULL,  -- {hot_state, tree_nodes, timestamp}
  updated_at INTEGER NOT NULL
);

Retention: 10 years (configurable) Cleanup: Monthly consolidation removes frozen entries older than retention period

Tree Index

Purpose: Hierarchical category map for O(log n) retrieval.

Constraints:

  • Max 50 nodes
  • Max depth 4 levels
  • Max 2KB serialized
  • Max 10 children per node

Example:

Memory Tree Index
==================================================
📂 Root (warm:15, cold:234)
  📁 owner — Owner profile and preferences
     Memories: warm=5, cold=89
  📁 projects — Active projects
     Memories: warm=8, cold=67
    📁 projects/evoclaw — EvoClaw framework
       Memories: warm=6, cold=45
      📁 projects/evoclaw/bsc — BSC integration
         Memories: warm=3, cold=12
  📁 technical — Technical setup and config
     Memories: warm=2, cold=34
  📁 lessons — Learned lessons and rules
     Memories: warm=0, cold=44

Nodes: 7/50
Size: 1842 / 2048 bytes

Operations:

  • --add PATH DESC
    — Add category node
  • --remove PATH
    — Remove node (only if no data)
  • --prune
    — Remove dead nodes (no activity in 60+ days)
  • --show
    — Pretty-print tree

Distillation Engine

Purpose: Three-stage compression of conversations.

Pipeline:

Raw conversation (500B)
  ↓ Stage 1→2: Extract structured info
Distilled fact (80B)
  ↓ Stage 2→3: Generate one-line summary
Core summary (20B)

Stage 1→2: Raw → Distilled

Input: Raw conversation text Output: Structured JSON

{
  "fact": "User decided to use raw JSON-RPC for BSC to avoid go-ethereum dependency",
  "emotion": "determined",
  "people": ["User"],
  "topics": ["blockchain", "architecture", "dependencies"],
  "actions": ["decided to use raw JSON-RPC", "avoid go-ethereum"],
  "outcome": "positive"
}

Modes:

  • rule
    : Regex/heuristic extraction (fast, no LLM)
  • llm
    : LLM-powered extraction (accurate, requires endpoint)

Usage:

# Rule-based (default)
distiller.py --text "Had a productive chat about the BSC integration..." --mode rule

# LLM-powered
distiller.py --text "..." --mode llm --llm-endpoint http://localhost:8080/complete

# With core summary
distiller.py --text "..." --mode rule --core-summary

Stage 2→3: Distilled → Core Summary

Purpose: One-line summary for tree index

Example:

Distilled: {
  "fact": "User decided raw JSON-RPC for BSC, no go-ethereum",
  "outcome": "positive"
}

Core summary: "BSC integration: raw JSON-RPC (no deps)"

Target: <30 bytes

LLM-Powered Tree Search

Purpose: Semantic search through tree structure using LLM reasoning.

How it works:

  1. Build prompt with tree structure + query
  2. LLM reasons about which categories are relevant
  3. Returns category paths with relevance scores
  4. Fetches memories from those categories

Example:

Query: "What did we decide about the hackathon deadline?"

Keyword search returns:

  • projects/evoclaw
    (0.8)
  • technical/deployment
    (0.4)

LLM search reasons:

  • projects/evoclaw/bsc
    (0.95) — "BSC integration for hackathon"
  • active_context/events
    (0.85) — "Deadline mentioned here"

LLM prompt template:

You are a memory retrieval system. Given a memory tree index and a query, 
identify which categories are relevant.

Memory Tree Index:
  projects/evoclaw — EvoClaw framework (warm:6, cold:45)
  projects/evoclaw/bsc — BSC integration (warm:3, cold:12)
  ...

User Query: What did we decide about the hackathon deadline?

Output (JSON):
[
  {"path": "projects/evoclaw/bsc", "relevance": 0.95, "reason": "BSC work for hackathon"},
  {"path": "active_context/events", "relevance": 0.85, "reason": "deadline tracking"}
]

Usage:

# Keyword search (fast)
tree_search.py --query "BSC integration" --tree-file memory-tree.json --mode keyword

# LLM search (accurate)
tree_search.py --query "what did we decide about hackathon?" \
  --tree-file memory-tree.json --mode llm --llm-endpoint http://localhost:8080/complete

# Generate prompt for external LLM
tree_search.py --query "..." --tree-file memory-tree.json \
  --mode llm --llm-prompt-file prompt.txt

Multi-Agent Support

Agent ID scoping — All operations support 

--agent-id
flag.

File layout:

memory/
  default/
    warm-memory.json
    memory-tree.json
    hot-memory-state.json
    metrics.json
  agent-2/
    warm-memory.json
    memory-tree.json
    ...
MEMORY.md              # default agent
MEMORY-agent-2.md      # agent-2

Cold storage: Agent-scoped queries via 

agent_id
column

Usage:

# Store for agent-2
memory_cli.py store --text "..." --category "..." --agent-id agent-2

# Retrieve for agent-2
memory_cli.py retrieve --query "..." --agent-id agent-2

# Consolidate agent-2
memory_cli.py consolidate --mode daily --agent-id agent-2

Consolidation Modes

Purpose: Periodic memory maintenance and optimization.

Quick (hourly)

  • Warm eviction (score-based)
  • Archive expired to cold
  • Recalculate all scores
  • Rebuild MEMORY.md

Daily

  • Everything in Quick
  • Tree prune (remove dead nodes, 60+ days no activity)

Monthly

  • Everything in Daily
  • Tree rebuild (LLM-powered restructuring, future)
  • Cold cleanup (delete frozen entries older than retention)

Full

  • Everything in Monthly
  • Full recalculation of all scores
  • Deep tree analysis
  • Generate consolidation report

Usage:

# Quick consolidation (default)
memory_cli.py consolidate

# Daily (run via cron)
memory_cli.py consolidate --mode daily

# Monthly (run via cron)
memory_cli.py consolidate --mode monthly --db-url "$TURSO_URL" --auth-token "$TURSO_TOKEN"

Recommended schedule:

  • Quick: Every 2-4 hours (heartbeat)
  • Daily: Midnight via cron
  • Monthly: 1st of month via cron

Critical Sync (Cloud-First)

Purpose: Cloud backup of hot state + tree after every conversation.

What syncs:

  • Hot memory state (identity, owner profile, active context, lessons)
  • Tree index (structure + counts)
  • Timestamp

Recovery: If device lost, restore from cloud in <2 minutes

Usage:

# Manual critical sync
memory_cli.py sync-critical --db-url "$TURSO_URL" --auth-token "$TURSO_TOKEN" --agent-id default

# Automatic: Call after every important conversation
# In agent code:
#   1. Process conversation
#   2. Store distilled facts
#   3. Call sync-critical

Retry strategy: Exponential backoff if cloud unreachable (5s, 10s, 20s, 40s)

Metrics & Observability

Tracked metrics:

{
  "tree_index_size_bytes": 1842,
  "tree_node_count": 37,
  "hot_memory_size_bytes": 4200,
  "warm_memory_count": 145,
  "warm_memory_size_kb": 38.2,
  "retrieval_count": 234,
  "evictions_today": 12,
  "reinforcements_today": 67,
  "consolidation_count": 8,
  "last_consolidation": 1707350400,
  "context_tokens_saved": 47800,
  "timestamp": "2026-02-10T14:30:00"
}

Usage:

memory_cli.py metrics --agent-id default

Key metrics:

  • context_tokens_saved — Estimated tokens saved vs. flat MEMORY.md
  • retrieval_count — How often memories are accessed
  • evictions_today — Memory pressure indicator
  • warm_memory_size_kb — Storage usage

Commands Reference

Store

memory_cli.py store --text "Fact text" --category "path/to/category" [--importance 0.8] [--agent-id default]

Importance guide:

  • 0.9-1.0
    — Critical decisions, credentials, core identity
  • 0.7-0.8
    — Project decisions, architecture, preferences
  • 0.5-0.6
    — General facts, daily events
  • 0.3-0.4
    — Casual mentions, low priority

Example:

memory_cli.py store \
  --text "Decided to deploy EvoClaw on BSC testnet before mainnet" \
  --category "projects/evoclaw/deployment" \
  --importance 0.85 \
  --db-url "$TURSO_URL" --auth-token "$TURSO_TOKEN"

# Store with explicit metadata (v2.1.0+)
memory_cli.py store \
  --text "Z-Image ComfyUI model for photorealistic images" \
  --category "tools/image-generation" \
  --importance 0.8 \
  --url "https://docs.comfy.org/tutorials/image/z-image/z-image" \
  --command "huggingface-cli download Tongyi-MAI/Z-Image" \
  --path "/home/user/models/"

Validate (v2.1.0)

memory_cli.py validate [--file PATH] [--agent-id default]

Purpose: Check daily notes for incomplete information (missing URLs, commands, next steps).

Example:

# Validate today's daily notes
memory_cli.py validate

# Validate specific file
memory_cli.py validate --file memory/2026-02-13.md

Output:

{
  "status": "warning",
  "warnings_count": 2,
  "warnings": [
    "Tool 'Z-Image' mentioned without URL/documentation link",
    "Action 'install' mentioned without command example"
  ],
  "suggestions": [
    "Add URLs for mentioned tools/services",
    "Include command examples for setup/installation steps",
    "Document next steps after decisions"
  ]
}

Extract Metadata (v2.1.0)

memory_cli.py extract-metadata --file PATH

Purpose: Extract structured metadata (URLs, commands, paths) from a file.

Example:

memory_cli.py extract-metadata --file memory/2026-02-13.md

Output:

{
  "file": "memory/2026-02-13.md",
  "metadata": {
    "urls": [
      "https://docs.comfy.org/tutorials/image/z-image/z-image",
      "https://github.com/Lightricks/LTX-Video"
    ],
    "commands": [
      "huggingface-cli download Tongyi-MAI/Z-Image",
      "git clone https://github.com/Lightricks/LTX-Video.git"
    ],
    "paths": [
      "/home/peter/ai-stack/comfyui/models",
      "./configs/ltx-video-2-config.yaml"
    ]
  },
  "summary": {
    "urls_count": 2,
    "commands_count": 2,
    "paths_count": 2
  }
}

Search by URL (v2.1.0)

memory_cli.py search-url --url FRAGMENT [--limit 5] [--agent-id default]

Purpose: Search facts by URL fragment.

Example:

# Find all facts with comfy.org URLs
memory_cli.py search-url --url "comfy.org"

# Find GitHub repos
memory_cli.py search-url --url "github.com" --limit 10

Output:

{
  "query": "comfy.org",
  "results_count": 1,
  "results": [
    {
      "id": "abc123",
      "text": "Z-Image ComfyUI model for photorealistic images",
      "category": "tools/image-generation",
      "metadata": {
        "urls": ["https://docs.comfy.org/tutorials/image/z-image/z-image"],
        "commands": ["huggingface-cli download Tongyi-MAI/Z-Image"],
        "paths": []
      }
    }
  ]
}

Retrieve

memory_cli.py retrieve --query "search query" [--limit 5] [--llm] [--llm-endpoint URL] [--agent-id default]

Modes:

  • Default: Keyword-based tree + warm + cold search
  • --llm
    : LLM-powered semantic tree search

Example:

# Keyword search
memory_cli.py retrieve --query "BSC deployment decision" --limit 5

# LLM search (more accurate)
memory_cli.py retrieve \
  --query "what did we decide about blockchain integration?" \
  --llm --llm-endpoint http://localhost:8080/complete \
  --db-url "$TURSO_URL" --auth-token "$TURSO_TOKEN"

Distill

memory_cli.py distill --text "raw conversation" [--llm] [--llm-endpoint URL]

Example:

# Rule-based distillation
memory_cli.py distill --text "User: Let's deploy to testnet first. Agent: Good idea, safer that way."

# LLM distillation
memory_cli.py distill \
  --text "Long conversation with nuance..." \
  --llm --llm-endpoint http://localhost:8080/complete

Output:

{
  "distilled": {
    "fact": "Decided to deploy to testnet before mainnet",
    "emotion": "cautious",
    "people": [],
    "topics": ["deployment", "testnet", "safety"],
    "actions": ["deploy to testnet"],
    "outcome": "positive"
  },
  "mode": "rule",
  "original_size": 87,
  "distilled_size": 156
}

Hot Memory

# Update hot state
memory_cli.py hot --update KEY JSON [--agent-id default]

# Rebuild MEMORY.md
memory_cli.py hot --rebuild [--agent-id default]

# Show current hot state
memory_cli.py hot [--agent-id default]

Keys:

  • identity
    — Agent/owner identity info
  • owner_profile
    — Owner preferences, personality
  • lesson
    — Add critical lesson
  • event
    — Add event to active context
  • task
    — Add task to active context
  • project
    — Add/update project

Examples:

# Update owner profile
memory_cli.py hot --update owner_profile '{"timezone": "Australia/Sydney", "work_hours": "9am-6pm"}'

# Add lesson
memory_cli.py hot --update lesson '{"text": "Always test on testnet first", "category": "blockchain", "importance": 0.9}'

# Add project
memory_cli.py hot --update project '{"name": "EvoClaw", "status": "Active", "description": "Self-evolving agent framework"}'

# Rebuild MEMORY.md
memory_cli.py hot --rebuild

Tree

# Show tree
memory_cli.py tree --show [--agent-id default]

# Add node
memory_cli.py tree --add "path/to/category" "Description" [--agent-id default]

# Remove node
memory_cli.py tree --remove "path/to/category" [--agent-id default]

# Prune dead nodes
memory_cli.py tree --prune [--agent-id default]

Examples:

# Add category
memory_cli.py tree --add "projects/evoclaw/bsc" "BSC blockchain integration"

# Remove empty category
memory_cli.py tree --remove "old/unused/path"

# Prune dead nodes (60+ days no activity)
memory_cli.py tree --prune

Cold Storage

# Initialize Turso tables
memory_cli.py cold --init --db-url URL --auth-token TOKEN

# Query cold storage
memory_cli.py cold --query "search term" [--limit 10] [--agent-id default] --db-url URL --auth-token TOKEN

Examples:

# Init tables (once)
memory_cli.py cold --init --db-url "https://your-db.turso.io" --auth-token "your-token"

# Query cold archive
memory_cli.py cold --query "blockchain decision" --limit 10 --db-url "$TURSO_URL" --auth-token "$TURSO_TOKEN"

Configuration

File: 

config.json
(optional, uses defaults if not present)

{
  "agent_id": "default",
  "hot": {
    "max_bytes": 5120,
    "max_lessons": 20,
    "max_events": 10,
    "max_tasks": 10
  },
  "warm": {
    "max_kb": 50,
    "retention_days": 30,
    "eviction_threshold": 0.3
  },
  "cold": {
    "backend": "turso",
    "retention_years": 10
  },
  "scoring": {
    "half_life_days": 30,
    "reinforcement_boost": 0.1
  },
  "tree": {
    "max_nodes": 50,
    "max_depth": 4,
    "max_size_bytes": 2048
  },
  "distillation": {
    "aggression": 0.7,
    "max_distilled_bytes": 100,
    "mode": "rule"
  },
  "consolidation": {
    "warm_eviction": "hourly",
    "tree_prune": "daily",
    "tree_rebuild": "monthly"
  }
}

Integration with OpenClaw Agents

After Conversation

import subprocess
import json

def process_conversation(user_message, agent_response, category="conversations"):
    # 1. Distill conversation
    text = f"User: {user_message}\nAgent: {agent_response}"
    result = subprocess.run(
        ["python3", "skills/tiered-memory/scripts/memory_cli.py", "distill", "--text", text],
        capture_output=True, text=True
    )
    distilled = json.loads(result.stdout)
    
    # 2. Determine importance
    importance = 0.7 if "decision" in distilled["distilled"]["outcome"] else 0.5
    
    # 3. Store
    subprocess.run([
        "python3", "skills/tiered-memory/scripts/memory_cli.py", "store",
        "--text", distilled["distilled"]["fact"],
        "--category", category,
        "--importance", str(importance),
        "--db-url", os.getenv("TURSO_URL"),
        "--auth-token", os.getenv("TURSO_TOKEN")
    ])
    
    # 4. Critical sync
    subprocess.run([
        "python3", "skills/tiered-memory/scripts/memory_cli.py", "sync-critical",
        "--db-url", os.getenv("TURSO_URL"),
        "--auth-token", os.getenv("TURSO_TOKEN")
    ])

Before Responding (Retrieval)

def get_relevant_context(query):
    result = subprocess.run(
        [
            "python3", "skills/tiered-memory/scripts/memory_cli.py", "retrieve",
            "--query", query,
            "--limit", "5",
            "--llm",
            "--llm-endpoint", "http://localhost:8080/complete",
            "--db-url", os.getenv("TURSO_URL"),
            "--auth-token", os.getenv("TURSO_TOKEN")
        ],
        capture_output=True, text=True
    )
    
    memories = json.loads(result.stdout)
    return "\n".join([f"- {m['text']}" for m in memories])

Heartbeat Consolidation

import schedule

# Hourly quick consolidation
schedule.every(2).hours.do(lambda: subprocess.run([
    "python3", "skills/tiered-memory/scripts/memory_cli.py", "consolidate",
    "--mode", "quick",
    "--db-url", os.getenv("TURSO_URL"),
    "--auth-token", os.getenv("TURSO_TOKEN")
]))

# Daily tree prune
schedule.every().day.at("00:00").do(lambda: subprocess.run([
    "python3", "skills/tiered-memory/scripts/memory_cli.py", "consolidate",
    "--mode", "daily",
    "--db-url", os.getenv("TURSO_URL"),
    "--auth-token", os.getenv("TURSO_TOKEN")
]))

# Monthly full consolidation
schedule.every().month.do(lambda: subprocess.run([
    "python3", "skills/tiered-memory/scripts/memory_cli.py", "consolidate",
    "--mode", "monthly",
    "--db-url", os.getenv("TURSO_URL"),
    "--auth-token", os.getenv("TURSO_TOKEN")
]))

LLM Integration

Model Recommendations

For Distillation & Tree Search:

  • Claude 3 Haiku (fast, cheap, excellent structure)
  • GPT-4o-mini (good balance)
  • Gemini 1.5 Flash (very fast)

For Tree Rebuilding:

  • Claude 3.5 Sonnet (better reasoning)
  • GPT-4o (strong planning)

Cost Optimization

  1. Use cheaper models for frequent operations (distill, search)
  2. Batch distillation — Queue conversations, distill in batch
  3. Cache tree prompts — Tree structure doesn't change often
  4. Skip LLM for simple — Use rule-based for short conversations

Example LLM Endpoint

from flask import Flask, request, jsonify

app = Flask(__name__)

@app.route("/complete", methods=["POST"])
def complete():
    data = request.json
    prompt = data["prompt"]
    
    # Call your LLM (OpenAI, Anthropic, local model, etc.)
    response = llm_client.complete(prompt)
    
    return jsonify({"text": response})

if __name__ == "__main__":
    app.run(port=8080)

Performance Characteristics

Context Size:

  • Hot: ~5KB (always loaded)
  • Tree: ~2KB (always loaded)
  • Retrieved: ~1-3KB per query
  • Total: ~8-15KB (constant, regardless of agent age)

Retrieval Speed:

  • Keyword: 10-20ms
  • LLM tree search: 300-600ms
  • Cold query: 50-100ms

5-Year Scenario:

  • Hot: Still 5KB (living document)
  • Warm: Last 30 days (~50KB)
  • Cold: ~50MB in Turso (compressed distilled facts)
  • Tree: Still 2KB (different nodes, same size)
  • Context per session: Same as day 1

Comparison with Alternatives

SystemMemory ModelScalingAccuracyCost
Flat MEMORY.mdLinear text❌ Months⚠️ Degrades❌ Linear
Vector RAGEmbeddings✅ Years⚠️ Similarity≠relevance⚠️ Moderate
EvoClaw TieredTree + tiers✅ Decades✅ Reasoning-based✅ Fixed

Why tree > vectors:

  • Accuracy: 98%+ vs. 70-80% (PageIndex benchmark)
  • Explainable: "Projects → EvoClaw → BSC" vs. "cosine 0.73"
  • Multi-hop: Natural vs. poor
  • False positives: Low vs. high

Troubleshooting

Tree size exceeding limit

# Prune dead nodes
memory_cli.py tree --prune

# Check which nodes are largest
memory_cli.py tree --show | grep "Memories:"

# Manually remove unused categories
memory_cli.py tree --remove "unused/category"

Warm memory filling up

# Run consolidation
memory_cli.py consolidate --mode daily --db-url "$TURSO_URL" --auth-token "$TURSO_TOKEN"

# Check stats
memory_cli.py metrics

# Lower eviction threshold (keeps less in warm)
# Edit config.json: "eviction_threshold": 0.4

Hot memory exceeding 5KB

# Hot auto-prunes, but check structure
memory_cli.py hot

# Remove old projects/tasks manually
memory_cli.py hot --update project '{"name": "OldProject", "status": "Completed"}'

# Rebuild to force pruning
memory_cli.py hot --rebuild

LLM search failing

# Fallback to keyword search (automatic)
memory_cli.py retrieve --query "..." --limit 5

# Test LLM endpoint
curl -X POST http://localhost:8080/complete -d '{"prompt": "test"}'

# Generate prompt for external testing
tree_search.py --query "..." --tree-file memory/memory-tree.json --mode llm --llm-prompt-file test.txt

Migration from v1.x

Backward compatible: Existing 

warm-memory.json
and 
memory-tree.json
files work as-is.

New files:

  • config.json
    (optional, uses defaults)
  • hot-memory-state.json
    (auto-created)
  • metrics.json
    (auto-created)

Steps:

  1. Update skill: 
    clawhub update tiered-memory
  2. Run consolidation to rebuild hot state: 
    memory_cli.py consolidate
  3. Initialize cold storage (optional): 
    memory_cli.py cold --init --db-url ... --auth-token ...
  4. Configure agent to use new commands (see Integration section)

Migration from v2.0 to v2.1

Fully backward compatible: Existing memory files work without changes.

What's new:

  • ✅ Metadata automatically extracted from existing facts when loaded
  • ✅ New commands: 
    validate
    , 
    extract-metadata
    , 
    search-url
  • ✅ 
    store
    command now accepts 
    --url
    , 
    --command
    , 
    --path
    flags
  • ✅ Distillation preserves URLs and technical details
  • ✅ No action required - just update and use new features

Testing the upgrade:

# Update skill
clawhub update tiered-memory

# Test metadata extraction
memory_cli.py extract-metadata --file memory/2026-02-13.md

# Validate your recent notes
memory_cli.py validate

# Search by URL
memory_cli.py search-url --url "github.com"

References

  • Design: 
    /docs/TIERED-MEMORY.md
    (EvoClaw)
  • Cloud Sync: 
    /docs/CLOUD-SYNC.md
    (EvoClaw)
  • Inspiration: PageIndex (tree-based retrieval)

v2.1.0 — A mind that remembers everything is as useless as one that remembers nothing. The art is knowing what to keep. Now with structured metadata to remember HOW, not just WHAT. 🧠🌲🔗