Security & Testing

mnemonik.xyz

Provides tools for creating, verifying, and recalling cryptographically signed memory attestations using Ed25519 signatures and blockchain anchoring.

1 endpoint7 known toolsFirst detected May 29, 2026Last detected September 6, 2026

ENDPOINT 1

https://mcp.mnemonik.xyz/mcp

No auth detected

Known tools 7

mnemonic_whoami

Returns this agent's cryptographic identity: Solana public key, did:sol, did:key, attestation count Purpose: Report on the local Mnemonic installation's health: which host configs contain a mnemonik entry, whether the hosted endpoint is reachable, binary integrity, local SQLite read/write, identity availability, and token availability.

Potential side effects

mnemonic_sign_memory

Creates a verifiable memory attestation: canonical CBOR + blake3 hash, signed with COSE_Sign1 (Ed25519), stored on Arweave, hash anchored as SPL Memo on Solana Purpose: Persist a verifiable, signed memory of a consequential decision or outcome so future agents (and the user) can recall it with cryptographic provenance.

Inferred read-only

mnemonic_verify

Verifies a memory attestation by recomputing hash and comparing against on-chain record Purpose: Confirm that a specific attestation is genuine: its content hash matches what is anchored on-chain, the COSE_Sign1 signature is valid against the claimed signer, and (for participate-mode rows) the Arweave and Solana records exist and agree.

Inferred read-only

mnemonic_prove_identity

Signs a challenge with Ed25519 key, proving identity without on-chain transaction Purpose: Guide the user through first-time setup so Mnemonic is wired into their MCP-capable agent host (Claude Code, Claude Desktop, Cursor) and the local pipeline is ready for offline attestations.

Inferred read-only

mnemonic_recall

Searches attested memory history using semantic similarity Purpose: Search the user's attested memory pool (or the public pool, if anonymous) by semantic similarity and return matching attestations ranked by relevance.

Inferred read-only

mnemonic_check_pending

Resolves a deferred-sign correlation_id to its on-chain state.

Inferred read-only

request_public_write_confirmation

Public-write ceremony gate: presents the content_hash about to be anchored on Arweave + Solana so the user can confirm or refuse in-turn before any chain write fires.

Potential side effects

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.mnemonik-xyz]
url = "https://mcp.mnemonik.xyz/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "mnemonik-xyz": {
      "type": "http",
      "url": "https://mcp.mnemonik.xyz/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: mnemonik-xyz
Remote MCP URL: https://mcp.mnemonik.xyz/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "mnemonik-xyz": {
      "url": "https://mcp.mnemonik.xyz/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "mnemonik-xyz": {
      "type": "http",
      "url": "https://mcp.mnemonik.xyz/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "mnemonik-xyz",
  "transport": "streamable-http",
  "url": "https://mcp.mnemonik.xyz/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.