Is this transaction safe to sign? Decodes what it really does and answers allow, warn or block.
MCPpedia last refreshed this data
io.github.godwin105/arbiter is an MCP server that is this transaction safe to sign? Decodes what it really does and answers allow, warn or block. Its tool list has not been published yet over sse and http, requires no API key, and scores 82/100 on MCPpedia's security, maintenance and efficiency rubric.
Config is the same across clients - only the file and path differ.
{
"mcpServers": {
"io-github-godwin105-arbiter": {
"args": [
"-y",
"@arbiterlabs/mcp"
],
"command": "npx"
}
}
}Are you the author?
Add this badge to your README to show your security score and help users find safe servers.
Is this transaction safe to sign? Decodes what it really does and answers allow, warn or block.
This server supports HTTP transport. Be the first to test it - help the community know if it works.
Five weighted categories - click any category to see the underlying evidence.
No known CVEs.
Checked @arbiterlabs/mcp against OSV.dev.
Click any tool to inspect its schema.
Be the first to review
Have you used this server?
Share your experience - it helps other developers decide.
Sign in to write a review.
Others in finance
OKX trading MCP server — connect AI agents to spot, swap, futures, options & grid bots via the Model Context Protocol.
Ledger of AI-agent money on Base: agents, payments, detections, Gate, signed Pass, free checks
Open-source AI Security Operations Center: alert fusion, LLM-agent triage, MITRE ATT&CK investigation, and a replayable decision ledger for every agent step. Self-hostable, runs with no API keys, MIT licensed. Ships an MCP server for Claude, Cursor and Continue.
Crypto briefs and data, Solana pre-trade checks, sanctions, papers, NLP. 102 tools, USDC, 20 chains.
MCP Security Weekly
Get CVE alerts and security updates for io.github.godwin105/arbiter and similar servers.
Start a conversation
Ask a question, share a tip, or report an issue.
Sign in to join the discussion.
The judgment layer for autonomous agents.
An AI agent about to act on the world often needs a judgment it cannot make by itself: is this transaction safe to sign, is this counterparty real, is this photo actually what it claims to be. Arbiter sells those judgments one call at a time, paid over x402 in USDC on Algorand.
Three routes, one response contract:
| Route | Judgment | Price |
|---|---|---|
POST /v1/judge/transaction | Machine — decode and risk-score an unsigned transaction before signing | $0.002 |
POST /v1/judge/counterparty | Data — is this payment counterparty who they claim to be | $0.01 |
POST /v1/judge/human | Human — questions a model cannot settle alone, answered by vetted reviewers | $0.25 |
Every route returns the same verdict envelope, so an agent integrates one shape and can then ask for any kind of judgment without new branching logic:
{
"decision": "block", // allow | warn | block | escalate
"risk": 100, // 0-100
"confidence": 1, // 0-1; low confidence forces "escalate"
"findings": [ /* itemised, each with severity, detail and source */ ],
"evidence": { /* the decoded transaction, the counterparty record */ },
"meta": { "degraded": false, "latencyMs": 378 }
}
New here, or explaining this to someone non-technical? What Arbiter actually does — no code, no jargon.
https://arbiter-hs23.onrender.com — API, and the reviewer app at
/work.
Preflight: 18 passed, 0 blocking failures. Running on Render's free plan, which has no persistent disk and spins down after 15 minutes idle — see DEPLOY.md for what that costs and the upgrade needed before the October leaderboard window.
| Component | State |
|---|---|
| x402 payment core (Algorand, USDC, GoPlausible facilitator) | Working, verified end-to-end |
Bazaar discovery + x402-global-challenge tag | Working, verified in the 402 response |
/v1/judge/transaction Algorand firewall | Working, 5/5 attack cases pass |
/v1/judge/transaction EVM firewall | Working, 7/7 drain patterns pass, EIP-7702 aware |
/v1/judge/counterparty engine | Working, 5/5 cases pass against live mainnet + NFD |
/v1/judge/human marketplace | Working, consensus + reviewer reliability + payout ledger |
| On-chain payout settlement | Working — real USDC paid to a reviewer on-chain |
| SDK, MCP server, LangChain, ElizaOS, paying proxy | Built, typechecked, runtime-verified |
| CrewAI tools (Python) | Built; runs through the paying proxy — see below |
| KeeperHub integration | Working, real Sepolia transaction executed |
| Reviewer app (React) | Live at /work, verified end-to-end with a paid question |
Autonomous agents lose funds in a small number of specific ways. The firewall decodes the actual transaction bytes and reports what signing would really do:
rekeyTo permanently transfers signing authority over the accountRun it against real, freshly-encoded transactions:
npm run demo:attacks
[BLOCK] Rekey attack risk=100/100 confidence=1 1365ms
[BLOCK] Close-remainder drain risk=100/100 confidence=1 378ms
[BLOCK] Asset close-out drain risk=100/100 confidence=1 1432ms
[BLOCK] Fee drain ris
... [View full README on GitHub](https://github.com/godwin105/arbiter_ai#readme)