The Agent Skills Directory

[SAFE]: No malicious patterns, such as obfuscation, hardcoded credentials, or unauthorized persistence, were found. The skill operates within its declared scope of blockchain interaction.
[COMMAND_EXECUTION]: The skill broadcasts transactions and calls functions on the Stacks blockchain, which is the primary intended functionality.
[DATA_EXFILTRATION]: The deploy command reads local files from user-specified paths using Bun.file().text() and broadcasts the contents on-chain. This is a functional requirement for contract deployment but constitutes a data ingestion and transmission path.
[PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection by ingesting file contents and JSON arguments.
Ingestion points: Contract source files in contract.ts via the --source argument.
Boundary markers: None; file contents are read directly into the transaction body.
Capability inventory: Subprocess execution for transaction broadcasting and file system access for reading contract code.
Sanitization: Contract names are validated via regex, and arguments are parsed using standard JSON utilities.

contract