Excited to be working with @3janexyz again to quantify risk and enable capital efficient credit onchain via LCC

Agents assessing agents, with results stored immutably on Base. That's the trust layer the agent economy needs. Full blog post → credprotocol.com/blog/mcp-se… Docs → docs.credprotocol.com/mcp-se… @anthropic @CoinbaseDev #MCP #ERC8004

Setup is one JSON block: { "mcpServers": { "cred-protocol": { "type": "url", "url": "api.credprotocol.com/mcp" } } } Then ask your agent: "What's the reputation of vitalik.eth?" It figures out the rest — which tools to call, how to interpret the results, what to flag.

What's in the toolkit: → 7 ERC-8004 agent reputation tools (discover, evaluate, and rate agents on-chain) → Sybil detection (0-100 risk score with behavioral indicators) → Identity attestations (ENS, Gitcoin Passport, POAPs, Worldcoin) → Reputation scoring (300-1000, like a credit score but fully on-chain) → Multi-chain activity reports across 10 EVM networks

The agent economy has a trust problem. Before your agent delegates a task, routes a payment, or grants access — it needs to know who it's dealing with. Is this wallet real? Is it a bot? Has this agent been rated well by other agents? That's what these tools answer.

We just shipped 21 MCP tools that give any AI agent instant access to on-chain reputation intelligence. Identity verification. Sybil detection. Agent-to-agent trust scoring. Financial profiling. All live, all real blockchain data. Connect in 2 minutes from Claude Code, Cursor, or Claude Desktop. 🧵

The full breakdown - how Stripe MPP crypto deposits work, a side-by-side comparison of all three payment paths, and what this means for agents building on paid APIs: credprotocol.com/blog/stripe…

Stripe is now processing machine payments on both sides: ◾ x402 — merchant settlement for API providers ◾ MPP — crypto deposits on Tempo One dashboard for all machine payment revenue, regardless of protocol. The settlement layer for the machine economy is taking shape.

Why three rails? Because agents aren't all the same: ◾ DeFi liquidation bot on Base → x402, minimal latency ◾ Enterprise compliance agent → Stripe, needs the audit trail ◾ Tempo-native agent → direct transfer, fastest path ◾ High-reputation agent (Cred Score 640+) → skips payment entirely, transacts on credit

Cred Protocol now supports 3 ways for agents to pay for on-chain credit data: → Direct Tempo — ~2s settlement, zero overhead → Stripe crypto deposit — ~30-60s, full Stripe dashboard + compliance → x402 via Coinbase CDP — permissionless, settles on Base or SKALE Same price ($0.01). Same data. Agent picks the rail.

Stripe now settles MPP payments on Tempo. We just added Stripe crypto deposits as a third payment rail for Cred Protocol's credit scoring API. An agent sends $0.01 USDC to a Stripe deposit address on Tempo → Stripe confirms → credit score delivered. No API key. No account. Just a wallet. 🧵👇

It's live right now. Try it: tempo request -X GET "api.credprotocol.com/api/v2/…" Accepts Tempo payments, Cred credit, x402, and traditional API keys. Full MPP spec compliance. 🔗 credprotocol.com 📄 docs.credprotocol.com ⚙️ mpp.dev

Reputation has teeth. Credit decisions are tied to ERC-8004 identity tokens. Every repayment and every default flows back on-chain. Default on a credit line? It tanks your Cred Score and locks you out of future credit across ANY service that checks it. Good actors build trust portably. Bad actors get priced out.

💸 tempo (direct pay) — any agent with a @tempo wallet pays $0.01 USDC per request. No account, no API key. 🏦 cred (pay on credit) — Cred Score ≥ 640? Transact on credit, no upfront payment. We check on-chain reputation and extend credit. Machines with good history pay later.

Cred Protocol is a reputation layer for autonomous agents. Your wallet's borrowing history, repayment behavior, and protocol interactions become a portable credit score any counterparty can query in real time. We wired it into the Machine Payments Protocol (MPP) with two modes 👇

🧵 We just shipped something at the Tempo hackathon that we've been thinking about for a while: Agentic reputation. AI agents are about to transact at scale. But right now every API call is prepaid — no identity, no history, no trust. Every agent starts at zero, every time.