The Agent Skills Directory

[DYNAMIC_EXECUTION]: The implementation guide includes a configuration loader that utilizes dynamic module loading to import environment-specific JSON files.
Evidence: The function loadConfig in references/implementation-guide.md uses require("../../config/documenso.${env}.json") where ${env} is derived from the NODE_ENV environment variable.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external environment files and local JSON configurations, which represents a surface for indirect instructions if those sources are compromised.
Ingestion points: The skill reads configuration from .env files and local JSON configuration files within the config/ directory.
Boundary markers: No explicit delimiter or "ignore instructions" markers are defined for the configuration data ingestion.
Capability inventory: The skill is configured with Read, Write, and Edit tool permissions.
Sanitization: The implementation guide demonstrates best practices by using zod schema validation to ensure the integrity of the loaded configuration data.

documenso-multi-env-setup