Settlement Predictor

Real-time on-chain settlement predictor for Ethereum, Bitcoin, Arbitrum, Optimism, Base & Polygon. Live gas tiers, mempool analysis, sandwich risk detection,...

This skill appears coherent with its stated purpose, but before installing or running it consider: 1) Network privacy — the tool uses public RPC endpoints (e.g., https://eth.llamarpc.com), mempool.space, and optional Etherscan/Tenderly APIs; queries you run (addresses, tx hashes, mempool scans) will be sent to those services. If you need privacy, replace endpoints with your own node or a trusted provider. 2) Local storage — it writes a cache/SQLite DB under ~/.cache/settlement-predictor (which can contain addresses/txs); inspect and remove the files if you don't want history retained. 3) Optional API keys — only provide ETHERSCAN_API_KEY or TENDERLY_API_KEY if you understand the privileges; don't export private keys to this environment. 4) Code review — the full Python file is bundled; if you have sensitive workflows, scan the code or run in an isolated/containerized environment. 5) Metadata mismatch — the manifest mentions gas-history.json while the code uses gas_history.db; this is likely harmless but shows the package hasn't been perfectly synchronized. If any of the above concerns are unacceptable, run the tool with your own RPC/mempool endpoints or avoid installing.

Type: OpenClaw Skill Name: settlement-predictor Version: 1.3.1 The settlement-predictor skill bundle is a legitimate tool for analyzing on-chain gas fees and predicting transaction settlement times for Ethereum, Bitcoin, and several Layer-2 networks. The Python script (settlement_predictor.py) uses standard libraries like web3 and requests to interact with public RPCs and official APIs (Etherscan, mempool.space, Tenderly, and CoinGecko). It implements local caching using a SQLite database in the user's home directory (~/.cache/settlement-predictor/), which is clearly documented in SKILL.md. The code is transparent, includes detailed error handling and timeouts, and lacks any indicators of malicious intent, data exfiltration, or prompt injection.