The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands via the Bash tool to identify changes in the repository, specifically git diff --name-only main...HEAD. This is an expected operation for the skill's stated purpose of reviewing code changes.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the codebase being reviewed, which creates a surface for indirect prompt injection.
Ingestion points: Content from files and file paths retrieved via Read, Grep, Glob, and git diff commands (SKILL.md).
Boundary markers: The skill utilizes a structured 6-aspect review framework and a mandatory 'Chain-of-Verification' protocol to validate findings, providing a logical structure for analysis, although explicit syntax delimiters for ingested code are not defined in the instructions.
Capability inventory: The skill has access to Bash, Read, Grep, and Glob tools, enabling it to interact with the local file system and execute shell commands.
Sanitization: No explicit sanitization or escaping of the code content is described before the agent processes it for review.

code-reviewer