Cross-Chain Arbitrage CN | 跨链套利

WarnAudited by ClawScan on May 10, 2026.

Overview

This skill is purpose-aligned for arbitrage research, but it defaults to executing crypto trades and bridges with unclear wallet limits and inconsistent provenance, so it needs careful review before use.

Treat this as a high-risk financial automation skill. Start with scan-only mode, connect only a dedicated limited-balance wallet, set an explicit maxAmount, verify the exact installed source, and do not allow execution unless you personally confirm each proposed transaction.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

High

#ASI02: Tool Misuse and Exploitation

What this means

If used in execute mode, the agent may bridge and trade funds across chains; stale prices, bad routing, or missing caps could cause financial loss.

Why it was flagged

The skill gives the agent an execution workflow for crypto arbitrage, makes execution the default, and does not set a default capital limit.

Skill content

| maxAmount     | No       | --              | Maximum capital to deploy per opportunity |
| mode          | No       | execute         | "execute" (full pipeline) or "scan" |
- Task(subagent_type:cross-chain-executor)

Recommendation

Use scan-only mode first, set a strict maxAmount, and require an explicit final approval for every trade or bridge transaction.

High

#ASI03: Identity and Privilege Abuse

What this means

A broadly authorized wallet or account could expose real funds to automated trading decisions.

Why it was flagged

The skill expects access to agent balances and capital-moving execution, but the supplied metadata declares no primary credential, env vars, or scoped account boundary.

Skill content

- mcp__uniswap__get_agent_balance

This skill automates the discovery and execution of cross-chain arbitrage opportunities

Recommendation

Use a dedicated low-balance wallet, verify exactly which Uniswap/MCP account is connected, limit allowances, and revoke permissions after use.

Low

#ASI04: Agentic Supply Chain Vulnerabilities

What this means

A user following the README could install a different or newer artifact than the one reviewed here.

Why it was flagged

The README points to unpinned external install sources and a different slug than the evaluated cross-chain-arbitrage-cn registry entry.

Skill content

npx skills add https://github.com/wpank/Agentic-Uniswap/tree/main/.ai/skills/cross-chain-arbitrage
...
npx clawhub@latest install cross-chain-arbitrage

Recommendation

Verify the publisher and exact repository/commit or registry slug before installing, especially because this skill can affect crypto funds.

Low

#ASI07: Insecure Inter-Agent Communication

What this means

Trading context and account balance information may be handled by external tools or subagents configured outside this artifact.

Why it was flagged

The workflow relies on multiple subagents and Uniswap MCP tools, which is purpose-aligned but means balances, opportunities, and execution context may cross tool/provider boundaries.

Skill content

- Task(subagent_type:opportunity-scanner)
- Task(subagent_type:risk-assessor)
- Task(subagent_type:cross-chain-executor)
- mcp__uniswap__check_safety_status
- mcp__uniswap__get_agent_balance

Recommendation

Confirm the trusted MCP server and subagent configuration before sharing wallet-connected context or allowing execution.