The Agent Skills Directory

[SAFE]: The skill is a standard developer tool designed for auditing code against documentation. It utilizes local file reading and code analysis utilities such as grep and the serena MCP tool.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes documentation and code files that could be influenced by external actors. Ingestion points: Reads tasks.md, design.md, implementation.md, and source code files. Boundary markers: The workflow does not explicitly define delimiters or instructions to ignore embedded commands when reading these files. Capability inventory: Limited to file reading and semantic analysis; the skill explicitly states that the agent should not implement changes without user confirmation. Sanitization: No explicit sanitization or filtering of documentation content is defined.

implementation-review