NeoGriffin Security API for AI Agents
"NeoGriffin is a multi-chain security API designed to protect AI agents in the OpenClaw ecosystem from the most critical threats in crypto: scam tokens, prompt injection attacks, malicious skill installations, and wallet drains. The API provides 10 security tools across free and paid tiers. Free tools include prompt injection scanning with 193 detection patterns and unicode evasion protection, NFT threat scanning for phishing and metadata injection, community-driven token reporting, and network statistics. Paid tools offer deep token auditing on Solana using Helius and DexScreener APIs, Base chain auditing via GoPlus and DexScreener, 24/7 wallet monitoring with 7 detection rules, and a Skill Supply Chain Scanner that analyzes OpenClaw skills for malicious code, obfuscation, credential theft, and hidden prompt injection before installation. NeoGriffin directly addresses five critical OpenClaw vulnerabilities: prompt injection, memory poisoning, credential exposure, excessive agency, and skill supply chain attacks. The payment system supports both SURGE token on Solana mainnet and USDC via the x402 micropayment protocol, making it accessible to both human users and autonomous agents. The server is live and tested with 9 successful mainnet payment verifications across all SURGE-enabled endpoints. All free endpoints have been validated with real blockchain data from Solana and Base networks."
Category tags:Additional links:.png&w=96&q=75){kind=small}
"Clean presentation professionally made and with good ordering of slides, however, it would have benefited from a voiceover to provide more context. It's also missing some key details for a 'pitch' deck for a product. The deck used in the presentation also made the assumption that whoever the audience is has prior knowledge of the domain. It should tell the story and explain the product in the simplest way possible. The deck provided separately (not in the video presentation) was easier to understand, but the deck in the video looked more professional. Somewhere in-between would have been better. Otherwise, I can see the value in the idea and it has potential. Some questions I might ask as an investor, which is generally applicable across all AI applications: - How can we be confident in the decisions / risk score? - What if it doesn't block an action it should have? - What if we block an action it shouldn't have? - Does it improve over time? Does it learn from itself? The presentation and deck would have also benefited from some stats around the problem would make it stronger. How much money has been lost as a result of these vulnerabilities? What impact does this tool have? e.g. Prevent X loss, etc. The 'market validation'. The presentation was also missing some key information: - What is the size of the market? What portion of this do we address? What is the size of the opportunity. - How do we make money from this? - How will you go to market? - Are there competitors in this space? What is your competitive advantage over them? There are some amazing pitch decks online you had read from big, successful companies that can help you think about how you tell the story of your product. For example, Airbnb. "
James Lloyd
Head of Digital Strategy
For Innovators & Creators