skills/bankrbot/skills/nexus/Gen Agent Trust Hub

nexus

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFECREDENTIALS_UNSAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill requires the agent to handle and transmit sensitive session credentials, specifically a wallet message signature (walletSig) and a bankrApiKey. These tokens authorize critical financial actions including trading, depositing, and withdrawing funds. The instructions include defensive constraints to prevent the agent from logging or persisting these secrets.
  • [DATA_EXFILTRATION]: High-privilege authentication data is transmitted to the vendor's remote infrastructure (og.nexustradinglabs.com) to execute authorized trades and manage collateral. This transmission is necessary for the skill's function but involves moving bearer credentials to a remote server.
  • [INDIRECT_PROMPT_INJECTION]: The agent is designed to ingest and process data from public trading feeds, trader profiles, and external news sources via an RSS proxy. This creates a surface for indirect prompt injection, where malicious instructions could be embedded in market data. The skill mitigates this with specific instructions for the agent to treat this data as untrusted and to never execute commands found within it.
  • [COMMAND_EXECUTION]: The installation instructions for the skill involve the use of npx, a command-line utility for executing Node.js packages, to add the skill to the environment. The skill also specifies a requirement for a bankr binary to be available in the execution environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 07:45 AM