MemoryMeshSAFE
A knowledge graph server that uses the Model Context Protocol (MCP) to provide structured memory persistence for AI models.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](./CHANGELOG.md) [](https://opensource.org/licenses/MIT)
MemoryMesh is a knowledge graph server designed for AI models, with a focus on text-based RPGs and interactive storytelling. It helps AI maintain consistent, structured memory across conversations, enabling richer and more dynamic interactions.
The project is based on the [Knowledge Graph Memory Server](https://github.com/modelcontextprotocol/servers/tree/main/src/memory) from the MCP servers repository and retains its core functionality.
[](https://mseep.ai/app/chemiguel23-memorymesh)
IMPORTANT
v0.3.0 Update: The MCP SDK has been updated from v1.0.4 to v1.25.2 to comply with the current Model Context Protocol specification (2025-11-25). This is a major update that brings compatibility with the latest MCP clients including Claude Desktop, ChatGPT, Cursor, Gemini, and VS Code. After updating, run npm install to fetch the new dependencies.
Since v0.2.7 the default location of schemas was changed to dist/data/schemas. This location is not expected to change in the future, but if you are updating from a previous version, make sure to move your schema files to the new location.
Quick Links
- Installation
- Example
- SchemaManager Guide Discussion
- [Memory
e8785b4d7adeOBSERVED · 2026-10-02Connect
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 memorymesh -- npx -y [email protected]
{
"mcpServers": {
"memorymesh": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (11)
3 read · 5 write · 3 destructive. Blast radius: 3 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
add_edges | write | Add multiple new edges between nodes in the knowledge graph. Edges should be in active voice |
add_metadata | write | Add new metadata to existing nodes in the knowledge graph |
add_nodes | write | Add multiple new nodes in the knowledge graph |
delete_edges | destructive | Delete multiple edges from the knowledge graph |
delete_metadata | destructive | Delete specific metadata from nodes in the knowledge graph |
delete_nodes | destructive | Delete multiple nodes and their associated edges from the knowledge graph |
open_nodes | read | Open specific nodes in the knowledge graph by their names |
read_graph | read | Read the entire knowledge graph |
search_nodes | read | Search for nodes in the knowledge graph based on a query |
update_edges | write | Update existing edges in the knowledge graph |
update_nodes | write | Update existing nodes in the knowledge graph |
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 (1 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (7)
delete_edges, delete_metadata, delete_nodes
import { EdgeWeightUtils } from '../../../src/core/graph/EdgeWeightUtils.js';import type { Edge } from '../../../src/core/graph/Edge.js';import { GraphValidator } from '../../../src/core/graph/GraphValidator.js';import type { Graph } from '../../../src/core/graph/Graph.js';import type { Node } from '../../../src/core/graph/Node.js';@modelcontextprotocol/sdk, zod, @types/node, @vitest/coverage-v8, copyfiles, shx, ts-node, tsc-alias
Gates applied: no_behavioural_pass.
e8785b4d7adefull audit observations/trust-audit/mcp-server/chemiguel23__memorymesh.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 |
|---|---|---|---|---|---|
| 2026-10-02 | e8785b4d7ade | SAFE | B | 89 | first audit |
Questions
What is the MemoryMesh MCP server?
A knowledge graph server that uses the Model Context Protocol (MCP) to provide structured memory persistence for AI models.
What tools does MemoryMesh expose?
11 in total: 3 read-only, 5 that write, and 3 that can delete or overwrite (delete_edges, delete_metadata, delete_nodes). Every one is listed on this page with its risk.
Is MemoryMesh 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: 3 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does MemoryMesh need?
No credential environment variables were found in its source, so it appears to need none.
How does MemoryMesh run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as memorymesh at 0.3.0.
How current is this page?
The grade is for one exact copy of the source (e8785b4d7ade), read on 2026-10-02. The repository is watched and re-audited when it changes.