The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes application and build logs through the railway logs command. These logs represent untrusted data that could contain malicious instructions designed to manipulate the agent's logic. * Ingestion points: railway logs command output. * Boundary markers: None specified in the instructions to separate log content from agent instructions. * Capability inventory: The skill can trigger service redeployments, restarts, and shutdowns. * Sanitization: No explicit filtering or sanitization of log content is provided.
[COMMAND_EXECUTION]: Several management commands are configured with the -y or --yes flags, which allow the agent to perform destructive or service-interrupting actions (redeploy, restart, stop) without requiring human confirmation. * Evidence: Use of -y in railway redeploy, railway restart, and railway down.
[METADATA_POISONING]: There is a discrepancy between the author name listed in the metadata ('Railway') and the provided author context ('davila7'), which may misrepresent the official nature of the skill. * Evidence: author: Railway field in SKILL.md frontmatter.

railway-deployment