The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell commands including git diff to identify changes, grep for technology detection, and mix for formatting, linting, and testing. These are standard operations for a development-focused code review agent.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted source code during the review process.\n
Ingestion points: Elixir source files (.ex, .exs, .heex) identified via git diff and grep (SKILL.md).\n
Boundary markers: Absent; the skill does not use specific delimiters or instructions to ignore prompts embedded in the code files.\n
Capability inventory: The skill can execute shell commands (mix), load additional skills, and spawn subagents via the Task tool (SKILL.md).\n
Sanitization: No validation or escaping of code content is performed before processing.

review-elixir