DVMCPSAFE
DVMCP is a bridge implementation that connects Model Context Protocol (MCP) servers to Nostr's Data Vending Machine (DVM) ecosystem
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Important Note: This project is archived. It has been superseeded by ContextVM/sdk.
DVMCP (Data Vending Machine Context Protocol) is a project that bridges the Model Context Protocol (MCP) with Nostr's Data Vending Machine (DVM) ecosystem. It enables AI and computational services running on MCP servers to be seamlessly discovered, accessed, and utilized via the decentralized Nostr network.
This integration combines MCP's standardized capability framework with Nostr's cryptographically secure and decentralized messaging, offering key advantages:
- Discoverability: MCP servers and their capabilities can be found through the Nostr network without centralized registries.
- Verifiability: Messages are cryptographically signed, ensuring authenticity and integrity.
- Decentralization: No single point of failure for service discovery or communication.
- Protocol Interoperability: Both MCP and DVMs leverage JSON-RPC patterns, facilitating smooth communication.
For a detailed technical specification, refer to the DVMCP Specification (2025-03-26).
Event Kinds
The following Nostr event kinds are defined and used within the DVMCP:
Packages
This monorepo contains the following packages:
@dvmcp/bridge
The bridge implementation that connects MCP servers to Nostr's DVM ecosystem. Handles tool an
fdc32fb0737cOBSERVED · 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 discovery -- npx -y @dvmcp/[email protected]
{
"mcpServers": {
"discovery": {
"command": "npx",
"args": [
"-y",
"@dvmcp/[email protected]"
]
}
}
}Exposed tools (8)
7 read · 0 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 |
|---|---|---|
discover | read | Discovers new tools with optional keyword filtering and tool integration. Use no keywords for broad discovery, or provide keywords as individual words or exact matches in quotes. Only integrate tools when explicitly required. |
input | read | Input text for the prompt |
invalid-tool | read | Invalid tool |
list_tools | read | Lists all currently registered tools, including built-in and discovered tools. |
remove_tool | destructive | Removes tools from the registry by ID, pattern, or author. |
test-echo | read | Echo test tool |
test-prompt | read | Test prompt for unit tests |
test-tool | read | A test tool |
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
- declared (3 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (7)
remove_tool
.rules
const tempConfigPath = join(import.meta.dir, '../../temp-test-config.yml');
@types/debug, @changesets/cli, @types/node, prettier
@getalby/lightning-tools, @modelcontextprotocol/sdk, @types/yargs, nostr-tools, yaml, yargs, @types/bun
@noble/hashes, debug, nostr-tools, yaml
@modelcontextprotocol/sdk, nostr-tools, yaml, yargs, @types/yargs, @types/bun
Gates applied: no_behavioural_pass.
fdc32fb0737cfull audit observations/trust-audit/mcp-server/gzuuus__dvmcp-1.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 | fdc32fb0737c | SAFE | B | 89 | first audit |
Questions
What is the DVMCP MCP server?
DVMCP is a bridge implementation that connects Model Context Protocol (MCP) servers to Nostr's Data Vending Machine (DVM) ecosystem
What tools does DVMCP expose?
8 in total: 7 read-only, 0 that write, and 1 that can delete or overwrite (remove_tool). Every one is listed on this page with its risk.
Is DVMCP 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 DVMCP need?
No credential environment variables were found in its source, so it appears to need none.
How does DVMCP run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as @dvmcp/discovery at 0.2.5.
How current is this page?
The grade is for one exact copy of the source (fdc32fb0737c), read on 2026-10-08. The repository is watched and re-audited when it changes.