meme-executor

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and consumes real-time price/liquidity data from public third-party services (e.g., "jupiter", "dexscreener", "birdeye" and RPCs like "helius"/"triton") as part of its pre-execution validation and execution pipeline (see the real_time_data and data_validation sections), exposing the agent to untrusted external content.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly an execution layer for Solana token swaps (memecoins). The prompt describes turning TradePlans into concrete orders, routing swaps via Jupiter/DEX router, wiring a Solana wallet private key / signer, and a roadmap that reaches "Mainnet live (small size)". It includes an explicit execute-trade component (currently DRY RUN) and a dryRun flag intended to be set false when wired to a real wallet. Although Phase 1 is simulation-only, the skill's primary and explicit design is to send on‑chain swap transactions (i.e., move crypto funds) once enabled. This matches the "Crypto/Blockchain (Wallets, Swaps, Signing)" criteria for Direct Financial Execution.