The Agent Skills Directory

[COMMAND_EXECUTION]: The SKILL.md file contains an example of using the grep command to search through the skill's reference documentation.
[DATA_EXFILTRATION]: The references/configuration-composition.md file provides code snippets for resolving environment variables using process.env. This pattern is used for application configuration but identifies a mechanism for accessing sensitive environment data.
[PROMPT_INJECTION]: The skill outlines patterns for dynamic configuration processing that present a surface for indirect injection if untrusted data is processed.
Ingestion points: The resolveEnvVars function in references/configuration-composition.md ingests configuration objects to resolve string tokens.
Boundary markers: The provided code templates do not include delimiters or instructions to ignore instructions embedded in the processed configuration.
Capability inventory: The documentation references capabilities for network operations (via axios and mqtt in references/factory-pattern.md) and command-line execution (grep in SKILL.md).
Sanitization: While the skill includes a validateConfig pattern, the environment variable resolution implementation does not provide built-in sanitization to restrict which environment variables can be accessed.

advanced-patterns