The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides a command reference for standard development tasks within the Astro monorepo, such as pnpm install, pnpm run build, and node --test. These commands are standard for project development and do not include any malicious parameters.
[EXTERNAL_DOWNLOADS]: The skill includes instructions to use pnpm install for dependency management. These dependencies are retrieved from the official npm registry, which is a well-known and trusted service.
[INDIRECT_PROMPT_INJECTION]: The skill requires the agent to read and interpret the Astro monorepo's codebase. This creates a surface for indirect prompt injection, as the agent's actions could be influenced by content within the repository. This is an inherent risk of development-oriented skills.
Ingestion points: Source code, configuration files, and documentation within the local withastro/astro monorepo.
Boundary markers: No explicit markers are used to delimit untrusted project content from agent instructions.
Capability inventory: Shell execution for pnpm and node commands, and filesystem write access for code changes.
Sanitization: The skill does not define specific sanitization routines for the data ingested from the repository.

astro-developer