The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands to execute project-defined scripts and standard tools such as npm, pnpm, npx, pyright, and ruff for building, linting, and testing code. These are standard operations for development workflows.
[PROMPT_INJECTION]: The skill processes untrusted project data through grep and git diff commands, creating an indirect prompt injection surface where malicious instructions in the codebase could attempt to influence the agent. * Ingestion points: Project source code and git metadata accessed during Phases 5 and 6. * Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the scanning process. * Capability inventory: The skill provides the ability to execute shell commands and read tool output into the agent context. * Sanitization: No explicit content validation or sanitization is performed on the data retrieved from the project files.

verification-loop