AtlasGraph
Back to Docs

Agent Platform

Tools for AI agent development. Transaction simulation, risk analysis, and code integrity verification.

Jump to: OverviewTX SimulationExecution TracerRisk ScoreAgent Verifier

Overview

Agent Platform provides 4 core tools to help AI agents interact safely with the blockchain.

TX Simulation

Simulate transactions before execution

Execution Tracer

Debug transaction execution step by step

Risk Score

Check risk score of contract addresses

Agent Verifier

Verify contract code integrity

TX Simulation

Execute transactions in a sandbox environment before sending to the blockchain. Predict failures, estimate gas costs, and detect MEV risks in advance.

What can you do?

  • Predict Failures — Check if a transaction will revert before sending
  • Gas Estimation — Get accurate gas usage predictions
  • MEV Risk Analysis — Detect sandwich attacks and frontrunning risks
  • Impact Analysis — Identify other contracts affected by the transaction

How to Use

  1. 1. Go to Agent Tools → TX Simulation in the sidebar
  2. 2. Select a chain and enter From, To addresses
  3. 3. Optionally enter Call Data and Value
  4. 4. Click Simulate
  5. 5. Review results: execution success/failure, gas used, MEV risks
Try TX Simulation

Execution Tracer

Debug transaction execution step by step. Trace internal calls, analyze failure causes, and get suggested fixes for reverted transactions.

What can you do?

  • Step-by-Step Trace — See every internal call (CALL, STATICCALL, DELEGATECALL)
  • Failure Analysis — Understand why a transaction reverted
  • Suggested Fixes — Get actionable recommendations to fix issues
  • Input/Output Data — View calldata and return data for each step

Two Modes

TX Hash Mode

Trace an existing transaction by hash. Analyze why it succeeded or failed.

Simulate Mode

Trace a simulated call without sending to chain. Debug before execution.

How to Use

  1. 1. Go to Agent Tools → Execution Tracer in the sidebar
  2. 2. Select TX Hash or Simulate mode
  3. 3. Enter transaction hash or call parameters
  4. 4. Click Trace
  5. 5. Review: execution steps, contracts involved, failure analysis
Try Execution Tracer

Risk Score

Evaluate the trustworthiness of any contract address with a 0-100 score. Check if a contract is risky before interacting with it.

What does it analyze?

  • On-chain Data — Contract age, verification status, activity level
  • ERC-8004 Reputation — Agent registration status, feedback score
  • Pattern Analysis — Phishing, rug pull, malicious code pattern detection

Risk Levels

Low

0 - 25

Medium

26 - 50

High

51 - 75

Critical

76 - 100

How to Use

  1. 1. Go to Agent Tools → Risk Score in the sidebar
  2. 2. Select a chain and enter the contract address
  3. 3. Click Check Risk
  4. 4. Review results: risk score, level, and detailed factors
Try Risk Score

Agent Verifier

Register a contract's bytecode hash and later verify if the code has been tampered with. Build trust between AI agents.

How does it work?

  • Register — Submit a contract address and we store its current bytecode hash
  • Verify — Compare current bytecode with the registered hash to detect changes
  • My Registrations — Manage your registered contracts

Verification Results

Match Hash matches — code unchanged
Mismatch Hash differs — code was modified
Not Registered Contract not found in registry

How to Use

Register a contract:

  1. 1. Go to Agent Tools → Verifier
  2. 2. Select the Register Code Hash tab
  3. 3. Enter chain and contract address, then click Register

Verify code integrity:

  1. 1. Select the Verify Code Hash tab
  2. 2. Enter chain and contract address, then click Verify
  3. 3. Review result: Match / Mismatch / Not Registered
Try Agent Verifier