← Back to Skills Marketplace
172
Downloads
0
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install settlement-predictor
Description
Real-time on-chain settlement predictor for Ethereum, Bitcoin, Arbitrum, Optimism, Base & Polygon. Live gas tiers, mempool analysis, sandwich risk detection,...
Usage Guidance
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.
Capability Analysis
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.
Capability Assessment
Purpose & Capability
Name/description, CLI examples, optional ETHERSCAN/TENDERLY keys, and the Python code all align: the tool fetches chain data, analyzes mempool/fees, predicts settlement, and optionally uses Etherscan/Tenderly. Requested/declared resources (no required env vars, optional API keys) are appropriate for this functionality.
Instruction Scope
SKILL.md instructs running the included Python CLI which performs network calls to public RPC endpoints, mempool.space, and (optionally) Etherscan/Tenderly. This is expected for the stated purpose, but the agent will contact third‑party services and may transmit transaction hashes, addresses, and mempool queries as part of normal operation.
Install Mechanism
No install spec is provided (instruction-only + bundled Python file). Dependencies are minimal (web3, requests). No remote downloads or execution of third‑party install scripts are present in the manifest.
Credentials
Only optional ETHERSCAN_API_KEY and TENDERLY_API_KEY are declared and documented; these map to features that legitimately require them (contract verification, simulation). No unrelated credentials are requested.
Persistence & Privilege
The skill creates local cache files under ~/.cache/settlement-predictor and uses a SQLite DB (DB path visible in code). That is reasonable for maintaining gas history, but the manifest's declared persistence path (~/.cache/settlement-predictor/gas-history.json) does not match the code's DB filename (~/.cache/settlement-predictor/gas_history.db) — an internal inconsistency. Local caches can contain query history (addresses, tx hashes) so users should be aware and may wish to audit or delete them.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install settlement-predictor - After installation, invoke the skill by name or use
/settlement-predictor - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.3.1
v1.3.1: Full rewrite (2450 lines) — all 15 commands implemented, SQLite cache, complete error handling. v1.3.0: Removed v2 roadmap section per user request.
v1.3.0
v1.3.0: Complete rebuild. Removed all hardcoded API keys (now env vars only). Moved history persistence to ~/.cache/. Added full credentials and persistence declarations to SKILL.md. Zero sensitive data in codebase.
v1.2.0
Fix: removed hardcoded API keys (all keys now optional via env vars). Fix: moved disk history from skill dir to ~/.cache/. Fix: added credentials declaration and persistence info to SKILL.md metadata.
v1.0.0
Initial release
Metadata
Frequently Asked Questions
What is Settlement Predictor?
Real-time on-chain settlement predictor for Ethereum, Bitcoin, Arbitrum, Optimism, Base & Polygon. Live gas tiers, mempool analysis, sandwich risk detection,... It is an AI Agent Skill for Claude Code / OpenClaw, with 172 downloads so far.
How do I install Settlement Predictor?
Run "/install settlement-predictor" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Settlement Predictor free?
Yes, Settlement Predictor is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Settlement Predictor support?
Settlement Predictor is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Settlement Predictor?
It is built and maintained by Ethan (@ethanwuqi-lang); the current version is v1.3.1.
More Skills