Nova MemorySAFE
AI Memory
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Local-first persistent memory for AI assistants. Store context across sessions with SQLite.
[](https://www.npmjs.com/package/@nova-mcp/mcp-nova) [](LICENSE)
What is Nova Memory?
Nova Memory is an MCP (Model Context Protocol) server that gives AI assistants persistent memory. It stores your project context, decisions, and knowledge locally in SQLite - no cloud, no telemetry. Your data stays on your machine.
Features
- 22 MCP tools for memory, search, analysis, and task management
- Store and search memories with BM25 full-text search
- Knowledge graph with relationship traversal
- Task management with phases and progress tracking
- Multi-project support with cross-project queries
- Enable/disable individual tools via configuration
- Complete TypeScript implementation with Zod validation
Installation
npm install -g @nova-mcp/mcp-nova
Setup
Claude Desktop
Add to your config file:
- Windows:
%APPDATA%\Claude\claude_desktop_config.json - macOS:
~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"nova-memory": {
"command": "npx",
"args": ["@nova-mcp/mcp-nova"]
}
}
}Claude Code
claude mcp add nova-memory -- npx @nova-mcp/mcp-nova
Available Tools
Memory Tools (8)
Search Tools (4)
ffa341c59646OBSERVED · 2026-10-07Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add mcp-nova -- npx -y @nova-mcp/[email protected]
{
"mcpServers": {
"mcp-nova": {
"command": "npx",
"args": [
"-y",
"@nova-mcp/[email protected]"
]
}
}
}Exposed tools (22)
20 read · 1 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
analysis_anomalies | read | Detect anomalies in memory patterns |
analysis_conflicts | read | Detect conflicts between memories |
analysis_forecast | read | Forecast memory trends and patterns |
analysis_memory_map | read | Generate visual map of memory relationships |
knowledge_graph | read | Build and query knowledge graphs from memories. Supports graph traversal, statistics, and multiple output formats (JSON, DOT, Mermaid). |
memory_compact | read | Compact memories to manage storage. Actions: status (check thresholds), preview (dry run), run (execute compaction) |
memory_delete | destructive | Delete a memory by ID |
memory_export | read | Export memories as JSON or markdown |
memory_import | write | Bulk import memories from JSON array |
memory_query | read | Query memories with filters and automatic pagination for large results |
memory_search | read | Search memories using semantic search with automatic pagination for large results |
memory_stats | read | Get memory statistics |
memory_tags | read | Get all tags with usage counts |
project_profile | read | Manage project-level memory profile. Categories: architecture, features, decisions, stack, conventions. |
search_patterns | read | Find patterns in memory sequences |
search_relationships | read | Find relationships between memories |
search_semantic | read | Perform semantic search across memories |
search_temporal | read | Search memories by time-based criteria |
store_memory | read | Store a memory entry with optional metadata |
task_board | read | Get current task board status with visual overview |
task_insights | read | Get AI-powered insights and analytics for tasks and workflow phases |
task_management | read | Comprehensive task and workflow management with phases, priorities, and smart insights |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (2 observation(s))
- Network
- declared (1 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (4)
memory_delete
return crypto.createHash("md5").update(normalized).digest("hex").substring(0, this.config.hashLength);const decoded = JSON.parse(atob(base64));
@babel/runtime, @modelcontextprotocol/sdk, @modelcontextprotocol/server-memory, @xenova/transformers, @zilliz/milvus2-sdk-node, blessed, chalk, chokidar
Gates applied: no_behavioural_pass.
ffa341c59646full audit observations/trust-audit/mcp-server/jagdeepsinghdev__nova-memory.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | ffa341c59646 | SAFE | B | 89 | first audit |
Questions
What is the Nova Memory MCP server?
AI Memory
What tools does Nova Memory expose?
22 in total: 20 read-only, 1 that write, and 1 that can delete or overwrite (memory_delete). Every one is listed on this page with its risk.
Is Nova Memory safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Nova Memory need?
No credential environment variables were found in its source, so it appears to need none.
How current is this page?
The grade is for one exact copy of the source (ffa341c59646), read on 2026-10-07. The repository is watched and re-audited when it changes.