Defi Arbitrage
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The package mostly matches an arbitrage agent but contains multiple mismatches and privacy/risk gaps (private-key handling, remote payment endpoint, CORS/logging), so it requires careful review before use.
This skill is plausible for DeFi arbitrage but contains inconsistencies and privacy risks. Before installing: (1) Do NOT send any real private keys to the agent or to POST /execute; the API requests executor_private_key which is sensitive. (2) Lock MERCHANT_BOT_URL to a trusted, local service (default is http://localhost:8202). If you must set it, ensure it points only to an endpoint you control; a remote merchant URL could be used to exfiltrate payments/metadata. (3) Run the agent in an isolated environment (container or VM) with restricted outbound network access except to your intended RPC endpoints. (4) Review the full source (there is a truncated portion in the provided package) to confirm whether any code actually sends keys or other data off-host; check for additional network calls and what is logged to agent.log. (5) Note packaging sloppiness: install.sh copies a missing Dockerfile, and skill.json and requirements.txt disagree on dependencies — ask the author for provenance and a reproducible build. If you need this capability and the author is trusted, run tests in a sandboxed environment and restrict MERCHANT_BOT_URL to localhost before providing any secrets.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
No visible risk-analysis findings were reported for this release.
