The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and audit external project files, which introduces an inherent surface for indirect prompt injection via malicious code comments or data within the projects being reviewed.
Ingestion points: Project source code files (.py), configuration files (pyproject.toml), and directory structures (apps/, packages/, src/).
Boundary markers: The instructions do not specify the use of clear delimiters or explicit "ignore embedded instructions" warnings for the agent when processing user-provided codebases.
Capability inventory: The skill primarily utilizes file-read and directory-listing capabilities. It also references development tools such as uv for environment synchronization.
Sanitization: The skill lacks explicit sanitization or filtering logic to prevent the agent from being influenced by instructions embedded in the project files under audit.
[SAFE]: The skill contains purely instructional content and architectural rules. There are no instances of network exfiltration, hardcoded credentials, or suspicious remote code downloads.

fastapi-rules