The Agent Skills Directory

[SAFE]: No security vulnerabilities, malicious patterns, or unsafe practices were detected. The skill instructions focus on standard development workflows.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it is designed to ingest and process external source code files. This is an inherent trait of code analysis tools and is considered a low-risk surface in this context.
Ingestion points: Source files targeted via the target or path parameters in /analyze:code, /dev:code-review, and /quality:improve (references/analyze-code.md, references/code-review.md, references/quality-improve.md).
Boundary markers: No explicit delimiters or 'ignore' instructions are used when passing file content to subagents.
Capability inventory: The agent has access to Bash, Read, Write, Edit, and MultiEdit tools across the provided reference files.
Sanitization: No explicit sanitization of ingested code content is mentioned.
[COMMAND_EXECUTION]: The skill mentions using the Bash tool for external analysis tools in references/analyze-code.md. This is a standard capability for analysis workflows and the skill explicitly states it will not execute dynamic analysis requiring compilation or runtime in its boundaries section.

code-quality-workflow