The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to read and process local project files, such as discovery.config.js and CMS component definitions in .jsonc format. * Ingestion points: Files under cms/faststore/components/ and the project configuration discovery.config.js are read to generate and synchronize store schemas. * Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions that might be embedded in the project data. * Capability inventory: The skill has the capability to execute shell commands using the vtex and faststore CLI tools, as well as the expect utility for automated interaction. * Sanitization: No explicit content sanitization or validation steps are described for the data ingested from the project files.
[COMMAND_EXECUTION]: The skill describes workflows involving the execution of developer-oriented CLI tools. * Evidence: SKILL.md and the reference documentation provide command sequences for project exploration (ls), content synchronization (faststore cms-sync), and manual schema uploads (vtex content upload-schema). It also includes a pattern for non-interactive prompt handling using the expect utility.
[EXTERNAL_DOWNLOADS]: The skill includes documentation and code examples that reference external network resources. * Evidence: references/extending-graphql-with-custom-resolvers.md contains examples of fetch calls to a public Brazilian postal code service (viacep.com.br) and VTEX-managed domains. These are documented as standard practices for extending the storefront's API functionality.

faststore-storefront