The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes common shell utilities such as grep, find, ls, and wc to perform audits of the project's source code for hardcoded style values and design system anti-patterns. These commands are executed locally within the project directories.\n- [EXTERNAL_DOWNLOADS]: A command is provided to allow the agent or user to submit feedback via npx --yes submit-expo-feedback@latest. This command downloads and executes a package from the NPM registry. This tool is managed by the vendor (Expo) and is used for its intended feedback purpose.\n- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data by instructing the agent to scan and read the content of the application's source code files. This presents a potential surface where malicious instructions embedded in the code could influence the agent's behavior.\n
Ingestion points: Application source code files located in directories such as app/, src/, components/, etc., are processed by audit commands in references/audit.md.\n
Boundary markers: No specific delimiters or instructions to ignore embedded content are used when processing the output of the audit commands.\n
Capability inventory: The skill has permissions to execute shell commands and write theme files to the project directory.\n
Sanitization: The output from the code scanning commands is not sanitized or filtered before being presented to the agent context.

expo-design-system