Miraix Binance Agent Firewall

AdvisoryAudited by Static analysis on Mar 16, 2026.

Overview

No suspicious patterns detected.

Findings (0)

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.

Note

ASI07: Insecure Inter-Agent Communication

What this means

A trading prompt may contain sensitive strategy or operational details, so that content would be shared with the Miraix API.

Why it was flagged

The skill explicitly sends the user's trading prompt and symbols to an external provider endpoint for scoring.

Skill content

curl -sS -X POST https://app.miraix.fun/api/binance-agent-firewall ... -d '{"prompt":"<raw-trading-prompt>","symbols":["<symbol-1>","<symbol-2>"]}'

Recommendation

Use only prompts you are comfortable sharing with the external service, and remove API keys, balances, personal identifiers, or private account information before auditing.

Note

ASI02: Tool Misuse and Exploitation

What this means

If a user asks for a Binance trading-prompt audit, the agent may call the Miraix endpoint as part of fulfilling that request.

Why it was flagged

The skill allows an agent to invoke it when the task appears to match, which may trigger the documented external audit workflow without the user manually selecting the skill each time.

Skill content

policy:
  allow_implicit_invocation: true

Recommendation

Confirm that external API use is acceptable before pasting sensitive trading prompts, and ask the agent not to call external services if you want a local-only review.