The Agent Skills Directory

[SAFE]: The skill consists of documentation and code snippets for interacting with the Xentral ERP REST API. It follows best practices by using environment variables for sensitive API tokens.
[DATA_EXFILTRATION]: The skill describes network operations using fetch to communicate with https://{instance}.xentral.biz/api. These operations are limited to the official API endpoints required for the skill's functionality.
[CREDENTIALS_UNSAFE]: The documentation and code examples correctly identify the need for Bearer token authentication and demonstrate using process.env.XENTRAL_API_TOKEN instead of hardcoding credentials.
[INDIRECT_PROMPT_INJECTION]: The skill outlines the process for ingesting and processing data from the external Xentral ERP system.
Ingestion points: JSON responses from Xentral API endpoints (customers, products, inventory, and orders).
Boundary markers: The skill does not provide specific boundary markers for the data returned from the API.
Capability inventory: The skill uses fetch for network communication and provides templates for CRUD operations on ERP resources.
Sanitization: The code snippets focus on API integration logic and do not explicitly detail data sanitization, which is expected for high-level integration guides.

xentral-erp-headless-checkout