General Tools

salt19.com

Provides general-purpose tools through the Model Context Protocol.

1 endpoint33 known toolsFirst detected September 13, 2026Last detected September 13, 2026

ENDPOINT 1

https://api.salt19.com/mcp

No auth detected

MCP server metadata

Name: salt19-agent-utility-grid
Version: 0.5.1

Capabilities

tools

Server instructions

Paid SALT19 tools use governed semantic idempotency. Generate one _salt19_operation_id per intended purchase, preserve it across the unpaid challenge and every retry, add _x402_payment_signature after satisfying the challenge, and never create a fresh operation ID merely because the prior outcome is ambiguous. Standard x402 clients may omit _salt19_operation_id on the initial challenge and paid retry; SALT19 derives a compatibility operation identity from the payment proof and canonical purchase. An explicit _salt19_operation_id is recommended for the strongest semantic at-most-once timeout and retry contract. SALT19 Payment Concierge exposes free x402-payment-preflight and report-payment-capability tools. The x402-authorization-handoff tool can mint a live challenge/retry template and report-x402-authorization-progress can cooperatively expose signer/signing stages. Buyers retain signing authority; SALT19 never requests a buyer private key or raw payment signature for telemetry.

Known tools 33

find-salt19-tool

FREE START-HERE ROUTER.

Inferred read-only

salt19-pricing

FREE catalog/pricing surface.

Inferred read-only

base-token-allowance

Read a live ERC-20 allowance on Base for an owner/spender pair without an API key.

Inferred read-only

base-token-balance

Read a live ERC-20 token balance on Base and return raw units plus best-effort token decimals/symbol metadata without signup or an API key.

Inferred read-only

base-transaction-status

Check current Base transaction receipt state and return PENDING, CONFIRMED_SUCCESS, or CONFIRMED_REVERTED with block and gas evidence.

Inferred read-only

base-wallet-balance

Read a wallet's live native ETH balance on Base at the latest block without an API key.

Inferred read-only

base-contract-state

Inspect whether a Base address currently has deployed bytecode and return code size plus best-effort ERC-20 metadata.

Inferred read-only

sec-recent-filings

Return recent SEC EDGAR filings for a company CIK from the live SEC submissions feed, normalized for agent consumption.

Inferred read-only

sec-company-facts

Return selected latest SEC XBRL company facts from EDGAR's live companyfacts feed with source dates and units.

Inferred read-only

nws-active-alerts

Return current U.

Inferred read-only

usgs-earthquakes

Query recent USGS earthquake events by lookback window and minimum magnitude and return compact live event state.

Inferred read-only

url-to-llm-markdown

Fetch a public web page, remove scripts/styles/navigation noise, and return bounded LLM-friendly Markdown plus canonical URL and fetch provenance.

Potential side effects

npm-package-facts

Return live npm registry facts for the latest or requested package version without the higher-priced vulnerability and adoption decision layer.

Inferred read-only

github-repo-facts

Return current public GitHub repository facts such as archive state, license, stars, forks, branch and latest push without the premium adoption decision.

Inferred read-only

agent-claim-output-verifier

Map an autonomous agent claim or output to supplied evidence and return SUPPORTED, PARTIALLY_SUPPORTED, NOT_PROVEN, or CONTRADICTED with explicit unsupported elements, contradictions, missing evidence, and recommended verification.

Inferred read-only

agent-code-change-risk-gate

Static pre-merge risk gate for autonomous code changes: inspect a supplied unified diff for security-sensitive surfaces, destructive operations, architecture/constraint conflicts, test requirements, rollback risk, and bounded breaking-change risk before an agent merges or deploys.

Inferred read-only

agent-decision-preflight

Before an autonomous coding agent installs or adopts software, return a bounded GO, WARN, or BLOCK decision with evidence and a portable SALT19 verification receipt.

Inferred read-only

agent-plan-verifier

Verify an autonomous agent plan before execution for missing preconditions, destructive steps, unsafe ordering, goal misalignment, constraint violations, and repair actions, returning a bounded GO/WARN/BLOCK result.

Inferred read-only

commerce-decision

Before an autonomous agent pays an x402 challenge, bind the request, accepted payment offer, and machine spending policy into a deterministic GO/WARN/BLOCK decision with SHA-256 commitments.

Potential side effects

commerce-receipt

Verify an x402 commercial state and persist a signed Ed25519 Proof402 receipt whose Commerce State Root commits request, offer, policy, payment, settlement, and delivery evidence; receipts are later Merkle-batched for public inclusion proofs.

Potential side effects

commerce-verify

Verify that an x402 v2 payment payload, settlement evidence, original request, accepted offer, spending policy, and delivered response are mutually bound; SALT19 operations can additionally be checked against authoritative D1 commercial state.

Potential side effects

dependency-go-no-go

Dependency approval decision for coding agents: decide whether an npm dependency or package is safe to install or adopt and return GO, WARN, or BLOCK using vulnerabilities, security, license, maintenance, deprecation, and package metadata.

Inferred read-only

dependency-upgrade-go-no-go

Compare current and target npm versions and return GO/WARN/BLOCK plus UPGRADE/DEFER/BLOCK guidance, bounded semver risk, OSV security implications, recommended tests, and explicit NOT_PROVEN runtime, peer, migration, and ecosystem compatibility boundaries.

Inferred read-only

github-repo-preflight

GitHub repository due diligence for coding agents: evaluate whether a public or open-source repository is safe and maintained enough to adopt using license, maintenance, release freshness, activity, project health, and risk signals.

Inferred read-only

model-memory-fit

Can this AI model run on my graphics card?

Inferred read-only

npm-package-intel

Check whether an npm package is deprecated or vulnerable before installing it, including vulnerabilities, license, maintenance, freshness, dependency health, and package.

Inferred read-only

repo-adoption-go-no-go

Repository adoption decision for autonomous coding agents using public GitHub maintenance, archive, license, and repository-health evidence, with explicit UNKNOWN and NOT_PROVEN boundaries for dependency, security, compatibility, integration, and project-fit claims.

Inferred read-only

url-http-preflight

Inspect a URL before an autonomous agent calls it, and preflight or verify a website or external API endpoint is healthy and safe to call by checking HTTP status, redirects, latency, and security headers.

Inferred read-only

verify-decision-receipt

Free integrity lookup for a SALT19 decision receipt.

Inferred read-only

x402-payment-preflight

Free buyer-side compatibility preflight.

Potential side effects

report-payment-capability

Free cooperative telemetry tool for reporting whether a client can use x402/Base/USDC or why it declined.

Potential side effects

x402-authorization-handoff

Free handoff tool that generates a live PAYMENT-REQUIRED challenge for a concrete paid request, returns exact REST and MCP retry templates, and creates a correlation ID.

Potential side effects

report-x402-authorization-progress

Free cooperative telemetry for the local buyer authorization path.

Inferred read-only

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.salt19-agent-utility-grid]
url = "https://api.salt19.com/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "salt19-agent-utility-grid": {
      "type": "http",
      "url": "https://api.salt19.com/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: salt19-agent-utility-grid
Remote MCP URL: https://api.salt19.com/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": {
    "salt19-agent-utility-grid": {
      "url": "https://api.salt19.com/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "salt19-agent-utility-grid": {
      "type": "http",
      "url": "https://api.salt19.com/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "salt19-agent-utility-grid",
  "transport": "streamable-http",
  "url": "https://api.salt19.com/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.