The Agent Skills Directory

[SAFE]: The skill exclusively utilizes official Unity Technologies resources. It references documentation from docs.unity.com and recommends official Unity packages such as com.unity.services.multiplayer and com.unity.netcode.gameobjects. These are trusted vendor resources provided by the skill author.
[EXTERNAL_DOWNLOADS]: The skill instructions direct the agent to fetch a documentation map from https://docs.unity.com/en-us/mps-sdk/llms.txt to ensure current and accurate API usage. This reference targets a well-known, official service and is used for context grounding rather than script execution.
[INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface by ingesting external data and workspace content. This is a standard requirement for its primary purpose of providing context-aware development assistance.
Ingestion points: The agent is instructed to read external documentation from docs.unity.com (as specified in SKILL.md) and inspect local project files including Packages/manifest.json and existing networking scripts (as specified in implementation-fit.md and workflows-prerequisites.md).
Boundary markers: None explicitly defined in the instructions for the ingested remote documentation.
Capability inventory: The skill provides logic for code generation and API usage advice. It does not define any shell execution, network-write, or file-system-write tools within its own configuration.
Sanitization: No specific sanitization or filtering of the ingested documentation or project metadata is mentioned.

setup-multiplayer-services