The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to manage the review workflow.
Evidence: Uses git diff --name-only HEAD~1 in SKILL.md to dynamically identify files for review.
Evidence: References a Read command to load context files from the .opencode directory.
[PROMPT_INJECTION]: The skill processes untrusted code files, which constitutes an indirect prompt injection surface.
Ingestion points: Code files identified via the $ARGUMENTS variable in SKILL.md.
Boundary markers: No explicit markers or instructions to ignore embedded commands are present in the processing logic.
Capability inventory: The skill can read local files and execute git commands as part of its core functionality in SKILL.md.
Sanitization: There is no evidence of sanitization or content validation for the files being reviewed.

code-review