All projects on Base matching “ai-tools”

Use when a Solana-side agent needs the current network state before sending a transaction, as often as it likes. Returns the current slot, block height and block time (unix and ISO) plus a prioritization-fee summary (min, median, p75, max in microlamports per compute unit, slots sampled, suggested fee = p75) from the public RPC, optionally scoped to up to five writable accounts. $0.002 per call vi

[MCP: optimize_context] Pay-per-call JSON context pruning & LLM token budget enforcement. Saves up to 85% prompt tokens.