The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied parameters such as trading symbols, quantities, and prices to construct shell commands, creating a potential surface for indirect prompt injection.
Ingestion points: The skill parses natural language instructions from the user to populate command arguments.
Boundary markers: The instructions mandate that a confirmation summary must be presented to the user to verify all parsed parameters before the command is executed.
Capability inventory: The skill is designed to execute shell commands using the revx CLI tool.
Sanitization: No explicit sanitization or validation logic is defined within the markdown instructions; the agent relies on the underlying CLI for input verification.
[SAFE]: The skill enforces a mandatory human-in-the-loop confirmation process for all high-risk actions, specifically for order placement, modification, and cancellation, which serves as a primary defense for real-world financial operations.
[SAFE]: Authentication management instructions are consistent with professional development practices, directing the agent to use a dedicated authentication skill (revx-auth) to resolve configuration issues while ensuring sensitive manual steps remain under user control.

revx-trading