The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes several grep commands in the verification section. These commands are designed to audit local Kotlin source files (*.kt) for specific implementation patterns. They use standard flags and are restricted to the local directory, which is a safe and common development practice for code analysis.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze data from the local codebase through grep. This creates an attack surface where maliciously crafted local source code could attempt to influence the agent's output. However, the skill lacks network access or file-writing capabilities to act on such instructions, and this behavior is essential to its primary purpose as a code auditing tool.
Ingestion points: Local .kt files via grep commands.
Boundary markers: None identified.
Capability inventory: Local file reading via grep shell execution.
Sanitization: None identified.

artwork-seeded-dynamic-scheme