enhanced-memory
Enhanced memory search with hybrid vector+keyword scoring, temporal routing, filepath scoring, adaptive weighting, pseudo-relevance feedback, salience scoring, and knowledge graph cross-references. Replaces the default memory search with a 4-signal fusion retrieval system. Use when searching memories, indexing memory files, building cross-references, or scoring memory salience. Requires Ollama with nomic-embed-text model.
Why use this skill?
Upgrade OpenClaw with 4-signal hybrid retrieval. Featuring temporal routing, salience scoring, and knowledge graph integration for superior memory recall.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/jameseball/enhanced-memoryWhat This Skill Does
The Enhanced Memory skill replaces OpenClaw's default flat vector search with a sophisticated 4-signal hybrid retrieval pipeline. By integrating vector similarity, keyword matching, header parsing, and filepath scoring, this tool achieves a significant boost in Mean Reciprocal Rank (MRR) to 0.782. It leverages the Ollama nomic-embed-text model to provide context-aware retrieval, handling temporal routing for date-sensitive queries and adaptive weighting for sparse search terms. The system also includes knowledge graph cross-referencing and autonomous salience scoring to identify stale or highly relevant information for agent self-improvement.
Installation
- Ensure Ollama is installed and run
ollama pull nomic-embed-text. - Install the skill via the ClawHub CLI:
clawhub install openclaw/skills/skills/jameseball/enhanced-memory. - Navigate to your workspace root and initialize the index:
python3 skills/enhanced-memory/scripts/embed_memories.py. - (Optional) Build the knowledge graph for improved entity correlation:
python3 skills/enhanced-memory/scripts/crossref_memories.py build.
Use Cases
- Deep Retrieval: When standard vector search fails to surface specific technical documentation or project notes.
- Temporal Context: Finding information based on relative dates like 'last Tuesday' or 'three weeks ago'.
- Knowledge Management: Automatically surfacing important but stale information during agent 'heartbeat' cycles.
- Complex Queries: When your request involves multiple topics where keyword matching is just as important as semantic meaning.
Example Prompts
- "Search my memories for the API key rotation schedule we discussed last Thursday."
- "What are the most salient, unaddressed tasks in my memory files based on the last month of activity?"
- "Find the design specification documents related to the authentication module and cross-reference them with my recent meeting notes."
Tips & Limitations
- Re-indexing: Always re-run
embed_memories.pyafter significant updates to your.mdfiles to ensure search accuracy. - Computational Cost: Because this skill uses a hybrid 4-signal fusion, it is more computationally intensive than basic vector search; monitor performance on resource-constrained devices.
- PRF: When the system triggers pseudo-relevance feedback, wait a moment longer for the result as it performs a secondary retrieval pass to expand query context.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-jameseball-enhanced-memory": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-read, file-write, code-execution