The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documentation in SKILL.md and references/lint-rules/README.md provides a Node.js script for project linting. It instructs users to manually review the code block, copy it to a local file, and execute it via the command line. This pattern involves the execution of code provided within the skill but relies entirely on manual user intervention and local execution, which is a standard practice for development tools.
[INDIRECT_PROMPT_INJECTION]: As a code review tool, the skill is designed to ingest and analyze untrusted data in the form of project source code. This represents a potential surface for indirect prompt injection if the analyzed code contains malicious instructions disguised as comments or data. However, the skill structures its analysis around specific rule-based 'LLM Check' prompts which focus the agent's attention on deterministic security and logic patterns, reducing the risk of accidental obedience to instructions embedded in the analyzed data.

cloudbase-code-review