The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a directory listing command (ls) to check for the presence of recommended companion skills in platform-specific hidden directories.
[EXTERNAL_DOWNLOADS]: The skill suggests the installation of additional companion components from the vendor's repository using the npx command, which involves remote code retrieval and execution.
[PROMPT_INJECTION]: The skill processes untrusted user-supplied contract text, creating a surface for indirect prompt injection.
Ingestion points: Contract drafts and drafting parameters provided by the user in the conversational context.
Boundary markers: Absent; the instructions do not define specific delimiters or instructions for the agent to distinguish contract data from control instructions.
Capability inventory: The agent is instructed to execute shell commands (ls) for environment discovery and provide installation recommendations.
Sanitization: No sanitization, escaping, or validation of the input contract content is mentioned or performed before processing.

contract-drafting