immunity-protocol

🧬 The Immunity SDK for AI agents. Sub-millisecond local cache, TEE-verified novel threats, on-chain settlement.

18
0
89% credibility
Found Apr 29, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A toolkit that lets AI agents safely check proposed blockchain actions against a shared network of crowd-sourced threat intelligence.

How It Works

1
📰 Discover Immunity

You learn about Immunity, a helpful network that keeps AI agents safe from scams and tricky actions.

2
📦 Add to your agent

You simply include the Immunity protector in your AI agent's setup.

3
💳 Link your wallet

Connect your digital money account so it can manage safe moves.

4
🌐 Join the safety network

Hook up to the shared group where agents warn each other about dangers.

5
🛡️ Start checking actions

Every time your agent wants to send money or interact, it gets scanned for threats.

6
Review the check
All clear

It's safe, so your agent goes ahead confidently.

Threat blocked

Matches a known bad pattern, keeping your funds secure.

⚠️
Double-check

Unsure case pauses for your quick yes or no.

7
📤 Share new threats

Spot a fresh danger? Share it to protect everyone else too.

🎉 Agent stays safe

Your AI agent now moves freely, protected by the whole community's wisdom.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is sdk?

Immunity SDK (github sdk typescript) equips AI agents with decentralized threat intelligence to block scams before they drain funds. Agents gate every on-chain action via a simple `check(tx, context)` call, hitting a sub-ms local cache of known threats first, falling back to on-chain registry lookup, then TEE-verified analysis of novel patterns using 0G Compute's qwen model. Hits settle on 0G Chain with USDC micropayments, while gossip propagates antibodies network-wide for collective defense.

Why is it gaining traction?

Its tiered lookup keeps checks under 200ms most times, avoiding TEE costs for known threats, unlike monolithic scanners that burn gas or latency. Economic incentives—stake USDC to publish, earn rewards on matches—turn users into publishers, creating a self-reinforcing intel network. The npm install + quickstart (mint test USDC, spin up AXL gossip mesh) hooks devs fast, with policies like "trust-cache" for production tuning.

Who should use this?

Builders of autonomous AI agents executing user txs, like wallet automators or DeFi bots prone to phishing/drainers. Teams securing langchain-style agents with tool calls or scraped sources, needing context-aware blocking without central honeypots.

Verdict

Grab it for testnet agents today—solid quickstart, integration tests, and Apache license make prototyping easy despite 18 stars signaling early days. Credibility score of 0.8999999761581421% flags risks, but strong docs and TEE gating mitigate; watch for mainnet maturity.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.