Plureslm Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Plureslm Mcp (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Distributed vector memory service for AI assistants powered by PluresDB.
PluresLM MCP provides persistent vector memory with P2P synchronization across devices. Built on PluresDB for distributed data and Model Context Protocol for AI tool integration.
npm install
npm run build
# Set PluresDB topic (generate with: openssl rand -hex 32)
export PLURES_DB_TOPIC="your-64-char-hex-topic-key"
# Start stdio MCP server
npm start# Configure for HTTP transport
export MCP_TRANSPORT=sse
export PORT=3001
export HOST=0.0.0.0
export PLURES_DB_TOPIC="your-topic-key"
# Start HTTP service
npm start
# → Serving on http://0.0.0.0:3001/sse| Variable | Required | Description |
|---|---|---|
PLURES_DB_TOPIC | ✅ | 64-char hex string (32 bytes) for PluresDB mesh |
PLURES_DB_SECRET | ❌ | Optional encryption secret for mesh |
MCP_TRANSPORT | ❌ | stdio (default) or sse for HTTP |
PORT | ❌ | HTTP port when using SSE transport (default: 3001) |
HOST | ❌ | HTTP host (default: 0.0.0.0) |
OPENAI_API_KEY | ❌ | OpenAI key for embeddings (falls back to local Transformers.js) |
OPENAI_EMBEDDING_MODEL | ❌ | OpenAI model name (default: text-embedding-3-small) |
PLURES_LM_DEBUG | ❌ | Enable debug logging (true/false) |
#### Local (stdio):
{
"mcpServers": {
"pluresLM": {
"command": "node",
"args": ["path/to/pluresLM-mcp/dist/index.js"],
"env": {
"PLURES_DB_TOPIC": "your-topic-key"
}
}
}
}#### Remote (SSE):
{
"mcpServers": {
"pluresLM": {
"transport": {
"type": "sse",
"url": "http://memory-service:3001/sse"
}
}
}
}PluresLM v2.0+ uses pure PluresDB for storage and synchronization:
All devices sharing the same PLURES_DB_TOPIC automatically sync memories:
# Device 1
export PLURES_DB_TOPIC="abc123..."
npm start # Stores memories locally
# Device 2
export PLURES_DB_TOPIC="abc123..." # Same topic
npm start # Automatically receives Device 1's memoriesPluresLM MCP exposes these tools for AI assistants:
pluresLM_store(content, tags?, category?, source?) - Store new memorypluresLM_search(query, limit?, minScore?) - Semantic searchpluresLM_forget(id? | query?, threshold?) - Delete memoriespluresLM_index(directory, maxFiles?, category?, tags?) - Index codebasepluresLM_status() - Database stats + sync statuspluresLM_profile() - User profile dataFor enterprise deployments across multiple OpenClaw instances:
# Memory service (dedicated server)
docker run -p 3001:3001 -e MCP_TRANSPORT=sse plures/pluresLM-mcp
# Worker instances (point to service)
export MCP_TRANSPORT=sse
export PLURES_LM_SERVICE_URL=http://memory-service:3001/sse# Multiple services with shared PluresDB topic
docker-compose up -d # Load balancer → N service instancesPluresLM v2.0 is a breaking change from SQLite-based v1.x:
memory_* → pluresLM_*)pluresLM_export_bundlepluresLM_import_bundleNote: Direct file migration not supported due to schema differences.
Dual-licensed under BSL-1.1 and MIT. You may choose either license at your option.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.