The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the commander library to define a CLI interface for blockchain operations. It executes commands such as deploy, call, and read. These involve reading local file content (via Bun.file().text()) and broadcasting transactions to the Stacks network. The command execution is scoped to the skill's primary purpose of smart contract interaction.
[INDIRECT_PROMPT_INJECTION]: The skill processes external contract source files (.clar) and function arguments provided as JSON. There is a surface for indirect prompt injection if an attacker-controlled contract or data is processed. However, the skill implements significant mitigations including:
Capability Inventory: Capabilities are limited to specific Stacks blockchain transactions via the @stacks/transactions library.
Boundary Markers: The CLI uses strict argument parsing and JSON validation for inputs like --args and --post-conditions.
Sanitization: Input values are converted into typed Clarity values (via parseArgToClarityValue), preventing raw string interpolation into sensitive contexts.
Safety Features: The skill defaults to PostConditionMode.Deny to prevent unauthorized token transfers during contract calls.

contract