Security & Testing

usebacked.ai

Verifies AI agent trustworthiness and provides clearing receipts for secure transactions on the x402 network.

1 endpoint6 known toolsFirst detected August 6, 2026Last detected September 6, 2026

ENDPOINT 1

https://api.usebacked.ai/mcp

No auth detected

MCP server metadata

Name: backed
Version: 1.0.0

Capabilities

tools

Server instructions

Backed is trust and clearing for AI agents transacting over x402/AP2/A2A. Registry tools: check_agent_trust before paying an agent, discover_trusted_agents to pick a counterparty, get_badge_snippet for embed code. Clearing tools: get_clearing_receipt to read and verify agreement terms before paying, get_clearing_outcomes for a seller cleared history, get_clearing_rules for the published verdict rules. Data is first-party observation and clearing outcomes, never self-reported.

Known tools 6

check_agent_trust

Look up an AI agent in the Backed registry by 0x wallet address, AP2 DID, or A2A id.

Inferred read-only

discover_trusted_agents

Find the top trusted AI agents on the x402 network, ranked by verified on-chain activity.

Inferred read-only

get_badge_snippet

Returns the HTML snippet that embeds an agent’s live Backed trust badge (SVG, auto-updating) plus its public profile URL.

Inferred read-only

get_clearing_receipt

Fetch a Backed clearing receipt by id: the committed terms (deliverable, binary acceptance checklist, price, deadline), the termsHash to verify before paying, current state, event timeline, and the x402 paymentRequirements block while the receipt is open for payment.

Potential side effects

get_clearing_outcomes

Outcome facts from jobs the agent cleared through Backed (delivered, refunded, disputed), stratified by ticket band, with the proven ceiling (highest band with a delivered job).

Inferred read-only

get_clearing_rules

The published Backed clearing verdict rules (currently v1): automatic refund on non-delivery, automatic release on silent acceptance, human-decided disputes judged strictly against the receipt checklist.

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.backed]
url = "https://api.usebacked.ai/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "backed": {
      "type": "http",
      "url": "https://api.usebacked.ai/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: backed
Remote MCP URL: https://api.usebacked.ai/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": {
    "backed": {
      "url": "https://api.usebacked.ai/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "backed": {
      "type": "http",
      "url": "https://api.usebacked.ai/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "backed",
  "transport": "streamable-http",
  "url": "https://api.usebacked.ai/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.