A sophisticated local cognitive memory system that utilizes FSRS-6 spaced repetition to provide AI agents with persistent, human-like recall capabilities.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install vestige
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install vestige using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Vestige Memory Skill?
Vestige is a cutting-edge memory enhancement for AI agents, grounded in over 130 years of cognitive research. It implements the FSRS-6 (Free Spaced Repetition Scheduler) algorithm, spreading activation, and synaptic tagging to manage information. Unlike standard vector databases, this tool for Openclaw Skills mimics biological memory where information is strengthened through use or naturally fades over time, ensuring the agent's context remains relevant and efficient.
By running 100% locally, Vestige provides a secure way for agents to maintain persistent recall across sessions. It bridges the gap between short-term conversation context and long-term knowledge, allowing developers to build more personalized and context-aware Openclaw Skills that remember user preferences, architectural patterns, and historical bug fixes without manual intervention.
Vestige Memory Skill Use Cases
- Maintaining persistent user preferences such as coding styles, UI themes, and language choices.
- Recording and retrieving architectural decisions and project-specific patterns during development.
- Storing solutions to complex bugs to prevent regression and repeated troubleshooting.
- Setting future triggers and reminders that the agent can recall based on conversation flow.
- Creating a long-term knowledge base of project context that survives session resets.
How Vestige Memory Skill Works
- Ingestion: The agent captures information through triggers like 'Remember this' or explicit ingest commands.
- Intelligent Processing: The smart_ingest tool performs duplicate detection and tags the information for semantic retrieval.
- Local Storage: Data is stored on the local filesystem with semantic embeddings for hybrid search capabilities.
- Spaced Repetition: The system applies the FSRS-6 algorithm, managing the 'strength' of memories based on how often they are accessed.
- Semantic Recall: Agents use the search tool to find relevant context via keyword, semantic, or hybrid queries at the start of new sessions.
- Reinforcement: Explicit promotion or demotion of memories further tunes the importance of stored data, ensuring high-value information stays at the forefront.
Vestige Memory Skill Setup
To integrate Vestige into your environment, ensure the binaries are located in your path. You can create a helper script for easier CLI access.
# Verify binary locations
ls ~/bin/vestige-mcp
ls ~/bin/vestige
# Create a helper script at ~/bin/vmem
cat <<EOF > ~/bin/vmem
#!/bin/bash
ACTION=\$1
shift
case \$ACTION in
search)
echo "{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"tools/call\",\"params\":{\"name\":\"search\",\"arguments\":{\"query\":\"\$*\"}}}" | ~/bin/vestige-mcp 2>/dev/null | jq -r '.result.content[0].text // .error.message'
;;
save)
echo "{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"tools/call\",\"params\":{\"name\":\"smart_ingest\",\"arguments\":{\"content\":\"\$*\"}}}" | ~/bin/vestige-mcp 2>/dev/null | jq -r '.result.content[0].text // .error.message'
;;
stats)
~/bin/vestige stats
;;
*)
echo "Usage: vmem [search|save|stats] [content]"
;;
esac
EOF
chmod +x ~/bin/vmem
Vestige Memory Skill Data Schema & Taxonomy
Vestige organizes data across specific system paths and utilizes a structured toolset for management:
Storage Locations
| OS | Path |
|---|---|
| macOS | ~/Library/Application Support/com.vestige.core/ |
| Linux | ~/.local/share/vestige/ |
| Cache | ~/Library/Caches/com.vestige.core/fastembed/ |
Primary MCP Tools
search: Performs hybrid semantic/keyword search.smart_ingest: Handles intelligent storage with deduplication.intention: Manages reminders and future triggers.promote_memory/demote_memory: Manually adjusts the FSRS-6 weight of specific data points.
Vestige Memory Skill Advanced Features
- FSRS-6 Spaced Repetition: Implements advanced memory decay and reinforcement logic to prioritize frequently needed information.
- Hybrid Semantic Search: Combines traditional keyword matching with deep semantic understanding for highly accurate retrieval within Openclaw Skills.
- Intention Triggers: Allows the agent to set proactive reminders that are triggered by context rather than just time.
- Local Embedding Cache: Optimizes performance by caching vector embeddings locally using fastembed.
- Cognitive State Management: Provides health checks and statistics to monitor the density and quality of the agent's memory bank.
SKILL.md
Loading
Related Openclaw Skills
Vynn Backtester
A natural language quantitative backtesting engine for trading strategy validation and financial analysis.
q-kdb-code-review
An AI-powered specialized code reviewer for Q/kdb+ designed to catch performance bottlenecks, type errors, and security vulnerabilities in high-frequency trading codebases.
Astrai Code Review
An intelligent AI code review agent that optimizes costs by routing diffs to the most efficient model based on complexity.
ClawPressor
ClawPressor is an intelligent session context compressor that uses NLP summarization to reduce token usage and extend session lifetime.
AKShare Finance
A comprehensive financial data interface library providing a unified wrapper for stocks, futures, crypto, and macroeconomic indicators.
Bilibili Downloader
A high-performance media extraction tool for Bilibili that supports 8K video downloads, audio extraction, and metadata retrieval.
Featured*
