The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes transaction data generated by a separate skill (swap-build), which acts as a vulnerability surface.
Ingestion points: The skill ingests transaction parameters including the target address, calldata, gas limits, and sender address from a JSON object.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the processing flow for the input data.
Capability inventory: The skill possesses the ability to execute shell commands via cast call and generate executable JavaScript code.
Sanitization: The skill implements a validation step that compares the provided router address against a hardcoded KyberSwap contract address, requiring manual user confirmation if they do not match.
[COMMAND_EXECUTION]: The skill instructs the agent to build and run a shell command (cast call) using several user-controlled variables (sender, value, gas, data, and to). If these values are not properly sanitized by the agent before shell execution, it could lead to command injection if the input source is compromised.
[DYNAMIC_EXECUTION]: The skill provides a template for generating and executing an ethers.js script. While using a standard library, the process involves runtime code generation based on external parameters.

swap-simulate