The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted user requirements to generate and potentially execute code, creating a surface for injection attacks.
Ingestion points: The user provides descriptions of "what to build" which are interpolated into code generation tasks.
Boundary markers: Absent. The instructions do not provide explicit delimiters or warnings to the agent to prevent the execution of instructions embedded within user requirements.
Capability inventory: The skill utilizes Write and Edit for file system modification and Bash (restricted to npm and node) for command execution in SKILL.md.
Sanitization: Absent. There are no instructions for the agent to sanitize or validate user-provided data before incorporating it into executable scripts or tools.
[COMMAND_EXECUTION]: The skill suggests several command-line operations for deployment and workflow management.
Evidence: The documentation recommends using npx gh-pages, netlify deploy, vercel --prod, and git worktree add to manage deliverables and parallel workstreams.
Context: These operations are consistent with the stated purpose of the skill (a "builder" tool) and target well-known deployment services (GitHub, Netlify, Vercel).

builder