The Agent Skills Directory

[SAFE]: The skill consists entirely of technical documentation and code examples intended to guide an AI agent in assisting users with the oRPC framework. The instructions are descriptive and focused on developer productivity without any malicious patterns, obfuscation, or attempts to bypass safety guardrails.
[EXTERNAL_DOWNLOADS]: The documentation includes instructions for users to install standard library packages via NPM (e.g., @orpc/server, @orpc/client) and to clone official playground repositories using npx degit middleapi/orpc. These are typical developer workflows for framework adoption and do not involve unauthorized or hidden remote code execution by the agent itself.
[INDIRECT_PROMPT_INJECTION]: As a documentation skill, it is designed to help the agent process and explain user-provided code related to oRPC. While this involves ingesting untrusted user content, the skill does not grant the agent any dangerous execution capabilities (such as shell access or file writing) that could be exploited via indirect injection. No specific boundary markers are defined, but the intended use case is primarily informational.

orpc-guide