flaip-launchpad

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The required workflow instructs the agent to read quick start guides and guides from https://flaip.xyz/agent-quickstart.en.md, https://flaip.xyz/agent-quickstart.md, and https://flaip.xyz/agent-skill.en.md when local references are absent, which constitutes reading low-risk external web pages.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill contains instructions and references to fetch runtime operational quickstart guides and full guides directly from remote URLs (https://flaip.xyz/agent-quickstart.en.md, https://flaip.xyz/agent-quickstart.md, https://flaip.xyz/agent-skill.en.md), making them unverifiable operational external dependencies from an unknown/third-party domain, which triggers a high severity finding.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (medium risk: 0.30). The skill operates a launchpad and trading platform on the BNB Chain, handling direct financial actions including token creation, trades, graduation claims, and refunds. It relies on the external binance-agentic-wallet skill to preview and sign transactions. Because transaction signing and authorization are delegated to the separate Binance Agentic Wallet execution layer (an independent validation mechanism), the direct operations are protected by qualifying independent access control.