Obsidian MemorySAFE
🧠 MCP server for storing AI memories as Markdown files with Obsidian graph visualization support
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
MCP server that stores AI memories as Markdown files for visualization in Obsidian's graph view.
About
This project is a modified version of Anthropic's memory server that has been adapted for Obsidian integration. The original server stored memories in JSON format, while this version stores them as individual Markdown files with Obsidian-compatible [[link]] syntax for graph visualization.
Key Changes from Original
- Storage Format: Changed from JSON to individual Markdown files
- Obsidian Integration: Added
[[link]]syntax for relations - YAML Frontmatter: Metadata stored in frontmatter instead of JSON
- File Structure: Each entity becomes a separate
.mdfile - Removed Features: Simplified to focus on Markdown-only storage
Features
- Markdown Storage: Individual
.mdfiles for each entity - Obsidian Integration: Uses
[[link]]syntax for graph visualization - Knowledge Graph: Store entities, relations, and observations
- Search Functionality: Query across all stored memories
- YAML Frontmatter: Metadata stored in frontmatter
Storage Format
Each entity is stored as an individual Markdown file with:
- YAML frontmatter for metadata (entityType, created, updated)
- Obsidian-compatible `[[links]]` for relations
- Organized sections for observations and relations
Example entity file (John_Doe.md):
--- entityType: person created: 2025-07-10 updated: 2025-07-10 --- # John Doe ## Observations - Works at Tech Corp - Expert in TypeScript - Lives in Tokyo ## Relations - [[Manager of::Alice Smith]] - [[Collaborates with::Bob Johnson]] - [[Located in::Tokyo Office]]
##
13fb0ea70662OBSERVED · 2026-10-08Connect
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 obsidian-memory-mcp -- npx -y [email protected]
{
"mcpServers": {
"obsidian-memory-mcp": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (9)
3 read · 3 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_observations | write | Add new observations to existing entities in the knowledge graph |
create_entities | write | Create multiple new entities in the knowledge graph |
create_relations | write | Create multiple new relations between entities in the knowledge graph. Relations should be in active voice |
delete_entities | destructive | Delete multiple entities and their associated relations from the knowledge graph |
delete_observations | destructive | Delete specific observations from entities in the knowledge graph |
delete_relations | destructive | Delete multiple relations 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 |
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 (3)
delete_entities, delete_observations, delete_relations
const hash = crypto.createHash('md5').update(name).digest('hex').substring(0, 8);gray-matter, @types/node, shx, typescript
Gates applied: no_behavioural_pass.
13fb0ea70662full audit observations/trust-audit/mcp-server/yunaga224__obsidian-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 |
|---|---|---|---|---|---|
| 2026-10-08 | 13fb0ea70662 | SAFE | B | 89 | first audit |
Questions
What is the Obsidian Memory MCP server?
🧠 MCP server for storing AI memories as Markdown files with Obsidian graph visualization support
What tools does Obsidian Memory expose?
9 in total: 3 read-only, 3 that write, and 3 that can delete or overwrite (delete_entities, delete_observations, delete_relations). Every one is listed on this page with its risk.
Is Obsidian 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: 3 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Obsidian Memory need?
No credential environment variables were found in its source, so it appears to need none.
How does Obsidian Memory run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as obsidian-memory-mcp at 1.0.0.
How current is this page?
The grade is for one exact copy of the source (13fb0ea70662), read on 2026-10-08. The repository is watched and re-audited when it changes.