The Agent Skills Directory

[SAFE]: No security issues detected.
[DATA_EXPOSURE]: The skill reads local repository data, including git diffs, commit logs, and documentation files (e.g., CODING_STANDARDS.md, CONTRIBUTING.md, and specification files). This data access is restricted to the local environment and is essential for the skill's code review functionality.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes external, potentially untrusted content such as code diffs and issue descriptions.
Ingestion points: Untrusted data enters the agent context via git diff output and external specification files fetched from issue trackers or local documentation paths.
Boundary markers: The prompts for the sub-agents do not implement explicit delimiters or instructions to ignore embedded commands within the analyzed diffs or specifications.
Capability inventory: The skill utilizes git commands and the Agent tool to spawn sub-agents with general-purpose capabilities.
Sanitization: There is no evidence of sanitization or filtering of the diff content or specification text before it is presented to the sub-agents for analysis.
Note: This surface is inherent to the primary purpose of a code review tool and does not indicate malicious intent.

review