The Agent Skills Directory

[SAFE]: The skill serves as a procedural guide for an AI agent to perform analysis and output structured reports. It defines strict rules for file path resolution, emphasizing that relative paths must be resolved against the project root rather than sensitive system directories.
[DATA_EXPOSURE]: While the skill involves reading target files and writing reports, it does not facilitate external data exfiltration or access sensitive system credentials. Its operations are confined to the project workspace and a designated temporary directory structure.
[INDIRECT_PROMPT_INJECTION]: As an analysis tool, the skill ingests untrusted data (target files for review). However, its instructions focus on structured reporting and analysis rather than execution of content found within those files. Standard agent safety guardrails effectively manage the low risk associated with processing external content for review.

multi-cli-review