The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use the grep command to perform recursive searches within project directories. These searches are used to identify patterns like hardcoded hex values (#[0-9a-fA-F]{6}), arbitrary Tailwind values (bg-[#), and non-semantic HTML elements (<div onClick). These commands are standard for code auditing and are constrained to the local project environment.
[INDIRECT_PROMPT_INJECTION]: This category is identified because the skill's primary function is to ingest and analyze untrusted project data, including source code files (.tsx, .jsx, .css) and documentation (READMEs).
Ingestion points: Source code and documentation files within the project directory (e.g., SKILL.md instructions to discover structure from documentation).
Boundary markers: Absent; the skill does not explicitly instruct the use of delimiters when reading file contents into context.
Capability inventory: The skill uses grep and standard file-reading capabilities. It lacks high-risk capabilities such as network exfiltration, arbitrary file-writing, or credential access.
Sanitization: No sanitization is performed on the ingested code or documentation before it is processed by the agent. However, due to the lack of dangerous capabilities, the risk of an indirect injection causing harm is negligible.

design-consistency-auditor