Agentforge
- Updated July 8, 2026
- acarchidi/agentforge
io.github.acarchidi/agentforge is a MCP server that adds DeFi safety checks—wallets, contract docs, approvals, and tx decode—for AI agents.
About
io.github.acarchidi/agentforge is a remote Model Context Protocol server that acts as a safety and comprehension layer when AI agents touch DeFi. Instead of blindly proposing transactions, an agent can verify wallet context, pull human-readable contract documentation, scrutinize token approvals, and decode pending transaction calldata so a developer can reject risky moves before they hit chain. It fits security for crypto-enabled products and integrations where you wire wallet tools into Claude Code or similar clients. Operationally it can also support post-incident understanding when decoding failed or suspicious txs. Agentforge is not a custody provider or exchange integration—it augments judgment. Connect via streamable HTTP to the published endpoint; review the GitHub repo for tool coverage and trust assumptions before production use.
- DeFi safety layer: wallet checks, contract docs, approvals, transaction decode
- Hosted streamable-http remote at agentforge-taupe.vercel.app/mcp (v1.3.0)
- No local npm install required—HTTP MCP endpoint for supported clients
- Oriented to AI agents interacting with smart contracts and wallets
Agentforge by the numbers
- Data as of Jul 9, 2026 (Skillselion catalog sync)
claude mcp add --transport http agentforge https://agentforge-taupe.vercel.app/mcpAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| Transport | HTTP |
|---|---|
| Auth | None |
| Last updated | July 8, 2026 |
| Repository | acarchidi/agentforge ↗ |
What it does
Give on-chain AI agents a DeFi safety layer: wallet checks, contract documentation lookup, approval review, and transaction decoding before signing.
Who is it for?
Best when you're shipping on-chain agents, trading bots, or wallet copilots and need guardrails beyond raw RPC calls.
Skip if: Non-crypto SaaS products, or teams that require fully offline, self-hosted security review with no third-party MCP remote.
What you get
You get structured safety and documentation signals before signing, so agent-driven DeFi workflows fail closed on sketchy txs.
- Pre-transaction wallet and approval context for agent proposals
- Decoded transaction and contract documentation signals for human review
By the numbers
- Server version 1.3.0
- Remote MCP URL: https://agentforge-taupe.vercel.app/mcp (streamable-http)
Recommended MCP Servers
1Claw VaultHSM-backed vault secrets for AI agents (JIT fetch) plus prompt-injection and threat scanning.2
1passwordMCP server for 1Password service accounts — tools and resources for vaults and credentials18
402sentinel McpAssess an x402 counterparty's risk BEFORE paying: allow/review/block, scored on-chain (Base).1
A2adependencyinspector McpRemote MCP for A2A dependency inspector MCP, structured receipts, audit logs, and reviewer-ready evi
A2a Governance Bridge McpA2A Governance Bridge MCP server.A2aidentitytoll McpRemote MCP for A2A caller identity, scope policy, verdict receipts, and audit history.How it compares
DeFi transaction safety MCP, not a general smart-contract audit firm or portfolio tracker skill.
FAQ
Who is io.github.acarchidi/agentforge for?
Developers building AI agents that interact with Ethereum-compatible DeFi protocols and want MCP-accessible wallet and transaction safety tooling.
When should I use io.github.acarchidi/agentforge?
Use it while shipping or operating wallet-connected agents whenever you review approvals, explain contracts, or decode transactions before user confirmation.
How do I add io.github.acarchidi/agentforge to my agent?
Register the streamable-http remote URL https://agentforge-taupe.vercel.app/mcp in your MCP client’s remote server config and follow acarchidi/agentforge setup notes for auth or limits if any.