The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git commands (git diff, git log, git rev-parse) to extract code changes and metadata for analysis. This is standard behavior for a code review tool and operates within the local repository context.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the repository (code diffs and specification files) and includes it in prompts sent to sub-agents, creating a surface for indirect prompt injection.
Ingestion points: SKILL.md describes reading git diff output, git log, and files within docs/, specs/, or .scratch/ folders, as well as external issue tracker data.
Boundary markers: The instructions do not define delimiters or specific "ignore embedded instructions" warnings when passing diff or specification content to the Standards or Spec sub-agents.
Capability inventory: The skill reads local files, executes git commands, and spawns sub-agents to process content.
Sanitization: There is no evidence of sanitization, filtering, or escaping applied to the retrieved code or documentation before it is interpolated into the sub-agent prompts.

code-review