myriad-prediction-markets
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes market information (titles, keywords, and metadata) sourced from the Myriad protocol and the BNB Chain. Malicious actors could potentially create prediction markets with names or descriptions designed to inject instructions and influence the agent's trading logic.
- Ingestion points: Market data retrieved via the
myriad markets listcommand and processed by the agent to make decisions. - Boundary markers: The instructions do not define delimiters or specific "ignore instructions" prompts for external market data.
- Capability inventory: The agent can execute trades (
myriad trade buy/sell), claim funds (myriad claim all), and manage wallet credentials. - Sanitization: No explicit filtering or sanitization of market titles or descriptions is mentioned.
- [COMMAND_EXECUTION]: The skill relies on the execution of specialized CLI tools (
@myriadmarkets/cliand@moonpay/cli) to perform blockchain interactions. These tools manage sensitive data, including private keys stored in environment variables (MYRIAD_PRIVATE_KEY) and local configuration files (~/.config/myriad/config.json). - [EXTERNAL_DOWNLOADS]: The skill requires the installation of external Node.js packages from the NPM registry to function. These tools are provided by the protocol vendor for the intended purpose of prediction market trading.
Audit Metadata