Pentagonal Mcp

pentagonal-ai/pentagonal

This is a blockchain development toolkit that connects Claude to an 8-agent AI system for smart contract work. It handles security audits, contract generation, and compilation across 8 different blockchain networks. You'd reach for this when you need to quickly prototype or audit contracts without manually juggling multiple chain-specific toolchains and security checkers. The multi-agent approach means you get several AI perspectives on potential vulnerabilities rather than a single pass. Based on the naming, it likely supports major EVM chains plus a few others, though the specific networks aren't listed in the available details. Most useful when you're doing cross-chain development or want automated security reviews before deployment.

tools: 7
auth: Secret needed
transport: STDIO
env var: 1
last updated: Apr 12, 2026

View on GitHub

Tools

7 tools

Public tool metadata for what this MCP can expose to an agent.

pentagonal_lookupOne-stop token intelligence. Returns price, market cap, ATH, 24h volume, transactions, holders, liquidity, LP lock status, security flags (honeypot, mintable, pausable, taxes), socials, and verified source code for any token by contract address. Supports Ethereum, Solana, Poly...3 params

One-stop token intelligence. Returns price, market cap, ATH, 24h volume, transactions, holders, liquidity, LP lock status, security flags (honeypot, mintable, pausable, taxes), socials, and verified source code for any token by contract address. Supports Ethereum, Solana, Poly...

Parameters* required

addressstring

Contract address (EVM 0x... or Solana base58)

chainstring

one of ethereum · polygon · arbitrum · base · optimism · bscdefault: ethereum

fieldsarray

Which sections to return. Use specific fields for faster queries.

pentagonal_auditRun a full 8-agent security pen test on a smart contract. Agents specialize in reentrancy, flash loans, access control, gas griefing, oracle manipulation, MEV/front-running, integer overflow, and economic exploits. Returns findings grouped by severity (Critical → High → Medium...3 params

Run a full 8-agent security pen test on a smart contract. Agents specialize in reentrancy, flash loans, access control, gas griefing, oracle manipulation, MEV/front-running, integer overflow, and economic exploits. Returns findings grouped by severity (Critical → High → Medium...

Parameters* required

chainstring

one of ethereum · polygon · arbitrum · base · optimism · bscdefault: ethereum

codestring

Solidity or Anchor/Rust smart contract source code

use_learned_rulesboolean

Apply accumulated security rules from past auditsdefault: true

pentagonal_generateGenerate a production-quality smart contract from a natural language description. Supports EVM (Solidity/OpenZeppelin) and Solana (Anchor/Rust programs or SPL token configs). Uses self-learning security rules from past audits.4 params

Generate a production-quality smart contract from a natural language description. Supports EVM (Solidity/OpenZeppelin) and Solana (Anchor/Rust programs or SPL token configs). Uses self-learning security rules from past audits.

Parameters* required

chainstring

one of ethereum · polygon · arbitrum · base · optimism · bscdefault: ethereum

promptstring

Natural language description of the contract

solana_typestring

Solana only: program (Anchor/Rust) or token (SPL config)one of program · token

use_learned_rulesboolean

default: true

pentagonal_fixFix a specific vulnerability in a smart contract while preserving all existing functionality. Call once per finding, starting with critical severity.3 params

Fix a specific vulnerability in a smart contract while preserving all existing functionality. Call once per finding, starting with critical severity.

Parameters* required

codestring

Smart contract source code containing the vulnerability

finding_descriptionstring

Description of the vulnerability and its impact

finding_titlestring

Title of the finding from the audit report

pentagonal_compileCompile Solidity source code to ABI, bytecode, constructor arguments, and gas estimates. Returns everything needed for deployment with Foundry, Hardhat, or cast.2 params

Compile Solidity source code to ABI, bytecode, constructor arguments, and gas estimates. Returns everything needed for deployment with Foundry, Hardhat, or cast.

Parameters* required

codestring

Solidity source code to compile

contract_namestring

Specific contract name to extract (defaults to first found)

pentagonal_rulesView the accumulated security rules knowledge base. Pentagonal learns from every audit and extracts generalizable rules. Check this to see the current state of the self-learning system.

View the accumulated security rules knowledge base. Pentagonal learns from every audit and extracts generalizable rules. Check this to see the current state of the self-learning system.

No parameter schema in public metadata yet.

pentagonal_chainsList all supported blockchains with chain IDs, explorer URLs, and deployment details.

List all supported blockchains with chain IDs, explorer URLs, and deployment details.

No parameter schema in public metadata yet.

Configuration

PENTAGONAL_API_KEYsecret

Pentagonal API key (optional — free tools work without it, paid tools accept x402 USDC on Base)

Readme

Pentagonal

AI-powered smart contract forge with an 8-agent adversarial security audit system.

Generate, audit, fix, and compile Solidity and Anchor/Rust contracts across 8 chains including Ethereum, Base, Polygon, Arbitrum, Optimism, BSC, Avalanche, and Solana. Includes token intelligence lookups with market data, holder analysis, and honeypot detection. Self-learning security rules engine that improves with every audit.

MCP Server

Install via npm:

npx pentagonal-mcp

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "pentagonal": {
      "command": "npx",
      "args": ["-y", "pentagonal-mcp"],
      "env": {
        "PENTAGONAL_KEY": "your-api-key"
      }
    }
  }
}

HTTP (Claude Code / Cursor / Windsurf)

{
  "mcpServers": {
    "pentagonal": {
      "type": "http",
      "url": "https://www.pentagonal.ai/api/mcp",
      "headers": {
        "x-pentagonal-api-key": "your-api-key"
      }
    }
  }
}

Tools

View the full README on GitHub

Details

Repository: github.com/pentagonal-ai/pentagonal
Categories: AI & LLM Tools, Security & Pentesting
Package: pentagonal-mcp
Transport: STDIO
Auth: Required
Registry: MCP registry, active

Related AI & LLM Tools MCP servers

View all

AgentTwinagenttwin/agenttwin-agent-toolsTest AI agent workflows against service twins with injected faults. Runs locally; no source upload.0stars

Mcp Serverarkheiaai/arkheia-mcpDetect fabrication and hallucination in any LLM output. 30+ models supported.0stars

Attestattestagents/install-attest-agentScan agent payment endpoints (x402, AP2, L402, MPP, 402) for a trust grade A-F before paying.0stars

Web Scraper to Markdown APIbr0ski777/web-scraper-x402Extract clean markdown from any URL. Removes boilerplate. For RAG pipelines. x402.0stars

Heartbeat Silence Checkdailyaiagents-cpu/dailyai-osDetects when workspace/system_memory/heartbeats.json mtime exceeds the silence threshold (default...0stars

Delx Commerce — Pay-per-Result APIs for Agentsio.github.davidmosiah/delx-mcp-a2a986 pay-per-result MCP/x402 APIs for web, data, Base, reliability and agent delivery.0stars