The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill facilitates the installation of the official Goldsky CLI by downloading and executing a script from the vendor's primary domain (https://goldsky.com). This is a standard and expected installation pattern for the developer toolset.
[COMMAND_EXECUTION]: The instructions guide the agent to perform local shell operations such as project authentication (goldsky login) and subgraph management (goldsky subgraph deploy). These actions are necessary for the skill's stated purpose of subgraph migration.
[PROMPT_INJECTION]: The skill processes user-supplied data as part of its migration workflow, presenting a surface for indirect prompt injection.
Ingestion points: User-provided GraphQL endpoints and IPFS deployment hashes are ingested in the migration command examples within SKILL.md.
Boundary markers: The skill does not implement explicit boundary markers or delimiters when interpolating user data into shell commands.
Capability inventory: The skill uses the goldsky CLI tool for resource deployment and management.
Sanitization: No explicit sanitization or input validation logic is described in the skill; it assumes the user provides valid identifiers. Given the developer-centric use case and the requirement for user authentication to deploy, the risk is categorized as safe.

subgraph-migrate