Skip to content

Add SafeAgent (Aigen-Protocol) - crypto/DeFi token safety MCP#149

Open
Aigen-Protocol wants to merge 1 commit into
Puliczek:mainfrom
Aigen-Protocol:add-aigen-safeagent
Open

Add SafeAgent (Aigen-Protocol) - crypto/DeFi token safety MCP#149
Aigen-Protocol wants to merge 1 commit into
Puliczek:mainfrom
Aigen-Protocol:add-aigen-safeagent

Conversation

@Aigen-Protocol
Copy link
Copy Markdown

Adds SafeAgent to the MCP Security Servers section.

What it is
MCP server providing crypto/DeFi token safety primitives for AI agents:

  • 27 scam pattern detection (Solidity source-code analysis)
  • Real DEX swap simulation for honeypot detection (not just code analysis)
  • 6 EVM chains: Ethereum, Base, Optimism, Arbitrum, Polygon, BSC
  • 38 MCP tools, free during beta, no auth required

Why it belongs in security MCPs
Crypto trading is one of the few domains where an LLM agent making a wrong call costs the user immediately measurable money. Every other agent in this list catches a vulnerability somewhere (code, network, identity); SafeAgent catches it at the swap call site via:

  • on-chain ERC-7913 safety oracle (Base 0x37b9e9B8789181f1AaaD1cD51A5f00A887fa9b8e)
  • SafeRouter contract (Base 0xF6EFc5D5902d1a0ce58D9ab1715Cf30f077D8f6e) that atomically reverts unsafe swaps with structured custom error data
  • HMAC-SHA256 signed webhook alerts on wallet score changes (the /watch endpoint)

Live

Closes #116

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant