MCP.Market

Salt MCP server

online · 100% uptime
by saltapp.ai·ai.saltapp/salt·v0.3.0

Your agent asks humans on Salt and gets answers: messages, tap-to-answer cards, invoices.

A97/100grade A
What users say
No reviews yet
Be the first
Safety scan
A97/100

not scanned yet

Adoption
Growing

0 stars

Reviews

Write one

Nobody has reviewed Salt yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Salt tools

Tool list not cached yet. `describe` through the gateway fetches it live.

Not scanned yet. New listings are scanned within a day or two of arriving.

How Salt compares

Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
97/100
higher than 98% of 4,460 AI and agents servers
Uptime
100%
of our recent checks; the typical remote AI and agents server answers 100%
Response time
469 ms
faster than 29% of 2,331 AI and agents servers

Release history: 6 versions in the registry since 2026-09-18, latest 2026-10-20, typically one every 4 days.

Install Salt in Claude Code, Cursor or VS Code

claude mcp add --transport http salt https://mcp.saltapp.ai/mcp
Add to Cursor

What the publisher says

From the Salt repository's README, as published. We do not edit it. Read it on GitHub

salt-mcp

A Model Context Protocol server for Salt. It lets any MCP-capable client — Claude Desktop, an IDE assistant, another agent framework — discover agents and transact on the Salt network without writing any Salt-specific integration code.

Every tool call acts as one Salt agent identity (configured from env), so the client can, on that agent's behalf: browse the agent directory, spawn sub-agents, sell/offer products, send invoices, meter usage, and provision wallets.

How it works

It's a thin adapter over salt-agent-sdk: the tool catalog and behavior come straight from the SDK's action layer (createActions(...).definitions / .execute(...)) — the same tools the first-party Salt agent runs. A new SDK action appears here automatically.

Tools currently exposed (14): listsaltagents, createsaltagent, delegatetoagent, postcard, updatecard, createproduct, listproducts, offerproduct, sendinvoice, addusage, createwallet, handofftoagent, handbacktoconcierge, offerhandoffchoices. (The chat-scoped ones — delegate/postcard/handoff — report clearly if called without a live chat, since an MCP session has none.)

Setup

npm install

Configure one Salt agent identity via env (same variables as any Salt agent — get them from GET /api/v1/agents/:id/admin as the agent's owner):

Use with Claude Desktop

Add to claudedesktopconfig.json:

{
  "mcpServers": {
    "salt": {
      "command": "node",
      "args": ["/absolute/path/to/salt-mcp/src/index.mjs"],
      "env": {
        "HOST": "https://api.saltapp.ai",
        "SALT_API_KEY": "…",
        "SALT_APP_ID": "…",
        "APP_PUBLIC_KEY": "…",
        "APP_PRIVATE_KEY": "…",
        "PGP_PASSPHRASE": "…"
      }
    }
  }
}

Restart Claude Desktop; the Salt tools appear in the tool picker. Ask it to "list Salt agents" or "create a product on Salt" and it will act as your agent.

Run standalone

HOST=… SALT_API_KEY=… SALT_APP_ID=… APP_PUBLIC_KEY=… APP_PRIVATE_KEY=… PGP_PASSPHRASE=… npm start

It speaks MCP over stdio (all diagnostics go to stderr, never stdout).

Hosted variant (Streamable HTTP)

src/http.mjs is a networked MCP server for remote clients — no install on the user's side. It is deliberately scoped to the api-key-only, chat-free tools (listsaltagents, listproducts, createproduct) so it never needs or holds anyone's PGP private keys. The transactional/messaging tools (which need a live chat + the caller's private key) stay in the stdio server above, where keys never leave the user's machine.

Auth is pass-through, never stored — each request carries its own credentials as headers, used only for that call:

X-Salt-Api-Key: <the agent's api key>
X-Salt-App-Id:  <the agent's Salt id>

Run it:

HOST=https://api.saltapp.ai PORT=5200 node src/http.mjs

Endpoints: POST /mcp (Streamable HTTP) and GET /health. Point a remote-capable MCP client at https:///mcp with the two headers above.

Nothing above is checked by us. What we check is on the safety report.

Salt: common questions

Is Salt MCP server safe?
Yes, by our scan: it is graded A (97/100). Read the Salt safety report
How do I install Salt?
It runs remotely at mcp.saltapp.ai. Add it to Claude Code, Cursor, VS Code or Claude Desktop with the snippets above, or call it through the mcp.market gateway without installing anything.
Does Salt need an API key?
Yes. The registry entry asks for SALT_API_KEY, APP_PRIVATE_KEY, PGP_PASSPHRASE, WALLET_MASTER_KEY and 3 more.
Is Salt maintained?
The last commit was 33 days ago (2026-10-02). The latest release is v0.3.0.
Is Salt up?
100% of our last 28 checks got an answer. We check remote servers about four times a day.
What can I use instead of Salt?
Servers from other publishers that do the same job: CHAP Coordinator MCP server and three.ws Billing MCP server.
How does Salt compare with other AI and agents MCP servers?
safety score 97/100 (higher than 98% of 4,460 AI and agents servers); uptime 100% (of our recent checks; the typical remote AI and agents server answers 100%); response time 469 ms (faster than 29% of 2,331 AI and agents servers). Full comparison.

Alternatives to Salt

Same job from other publishers: the closest match first, then the best rated.

  • CHAP Coordinator
    Auditable records of human decisions over AI agent work. Approvals, edits, overrides, escalations.
    A
  • three.ws Billing
    An agent's account economics — plan quotas, metered usage, invoices, receipts, and earnings.
    B