[FEATURE] Add AgentTollSafetyTool: honeypot/rug checks for a Base token via x402

Feature Area

Integration with external tools

Is your feature request related to an existing bug? Please link it here.

NA

Describe the solution you'd like

Add AgentTollSafetyTool, a BaseTool wrapping AgentToll's /api/base/safety endpoint: a simulated buy and sell, buy/sell tax, owner privileges, holder concentration, liquidity risk, and the deployer's own history for a Base (chain id 8453) token contract.

AgentToll is a live, open-source (MIT), pay-per-call API settled in USDC on Base via the x402 protocol (HTTP 402) — no API key, no subscription, no signup. This gives CrewAI agents a real safety verdict on an unfamiliar Base token without any onboarding step, paid only on a successful response.

Describe alternatives you've considered

None — there's currently no tool in crewai_tools for checking token safety/honeypot risk on Base, and no x402-paid tool at all.

Additional context

A working implementation already exists and is verified (tests, ruff, mypy, a real import through the full crewai_tools package, and the exact x402 Python API exercised against its own upstream example): see PR #7226, which the first-time-contributor bot closed pending an associated issue. Reopening it and linking here as soon as this issue is created.

Willingness to Contribute

Yes, I'd be happy to submit a pull request

Start by reviewing PR #7226, which contains the working implementation, and inspect the AgentToll /api/base/safety endpoint and x402 Python API usage described in the issue. Check the existing tests, ruff and mypy results, full crewai_tools import, and upstream example verification. Done means the AgentTollSafetyTool is integrated and its Base-token safety checks work through the paid endpoint.