The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze and validate user-provided source code using specific validation rules. This capability creates an indirect prompt injection surface where malicious instructions could be embedded in the code under review to influence the agent's behavior.
Ingestion points: User-supplied source code files (e.g., src/**/*.ts, lib/**/*.js) processed during code review tasks defined in references/validations.md.
Boundary markers: The skill instructions in SKILL.md do not include explicit delimiters or warnings to treat processed code as untrusted content or to ignore embedded instructions within that code.
Capability inventory: The skill facilitates code analysis and provides architectural advice. While it does not include its own executable scripts, the agent using the skill typically has access to shell environments and file system tools.
Sanitization: There are no sanitization, escaping, or validation mechanisms defined for the input code before it is interpreted by the agent for analysis.
[SAFE]: The external Node.js dependencies and services referenced in the code patterns (such as pino, winston, uuid, and axiomhq) are well-known technology components used as intended for logging and observability.

logging-strategies