The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill's operations align with its documented purpose of infrastructure management.
[CREDENTIALS_UNSAFE]: The skill facilitates the management of sensitive environment variables.
Evidence: The list-variables and set-variables tools (references/api_reference.md) are designed to interact with service secrets such as DATABASE_URL and *_API_KEY.
[COMMAND_EXECUTION]: The skill orchestrates application deployment and project lifecycle.
Evidence: Includes tools for triggering deployments (deploy), creating projects (create-project-and-link), and modifying environment configurations (set-variables, create-environment) in references/api_reference.md.
[EXTERNAL_DOWNLOADS]: The skill references official external resources for setup.
Evidence: Provides links to https://docs.railway.com for CLI installation and documentation in SKILL.md.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through log ingestion.
Ingestion points: get-logs (references/api_reference.md) retrieves build and deployment logs which may contain untrusted data.
Boundary markers: Absent; the skill does not define delimiters or provide instructions to ignore embedded content within logs.
Capability inventory: The skill has tools to modify environment variables (set-variables) and trigger code deployments (deploy) in references/api_reference.md.
Sanitization: Absent; no content validation or escaping is performed on log data before it is presented to the agent.

railway