The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by retrieving untrusted data from remote CMS instances.
Ingestion points: Untrusted content is ingested into the agent context through npx dineway content get, npx dineway content list, and npx dineway search commands.
Boundary markers: The guidelines do not provide instructions for using delimiters or boundary markers to isolate remote content from system instructions.
Capability inventory: The agent has access to powerful tools including file system modification (init, seed), network requests to arbitrary URLs, and deployment orchestration (dineway deploy).
Sanitization: While the tool converts data formats (Portable Text to Markdown), it lacks sanitization logic for instructions that may be embedded in the fetched content.
[EXTERNAL_DOWNLOADS]: The skill relies on external code execution through standard package managers and CLI tools.
Vendor resources: It executes npx dineway, which downloads and runs the vendor's CLI from the npm registry.
Third-party CLIs: The dineway deploy command interacts with and may prompt for the installation of well-known service CLIs such as Railway (@railway/cli), Fly.io (flyctl), and Google Cloud (gcloud).
[COMMAND_EXECUTION]: The skill is primarily a set of instructions for shell command execution to perform administrative tasks, database operations, and content management.

dineway-cli