myriad-prediction-markets
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill processes external market data retrieved via the
myriad markets listcommand. This represents a surface for indirect prompt injection where maliciously crafted market metadata could attempt to influence the agent's behavior. - Ingestion points: Market titles, descriptions, and metadata retrieved from the Myriad Protocol API.
- Boundary markers: No delimiters or explicit instructions to ignore embedded content are present in the command patterns.
- Capability inventory: Trade execution (
myriad trade), winnings claims (myriad claim), and wallet configuration (myriad wallet). - Sanitization: No sanitization of market-provided strings is documented.
- [COMMAND_EXECUTION]: The skill makes extensive use of CLI tools (
myriad,mp) for core operations including wallet setup, trading, and running an MCP server. It also includes a command to modify agent environments (myriad skills install). - [EXTERNAL_DOWNLOADS]: The skill requires the global installation of the
@myriadmarkets/cliand@moonpay/clipackages from the npm registry to function.
Audit Metadata