dominodatalab/domino-ai-plugin / skills/domino-llm-gateway
domino-llm-gateway
VerifiedCall Large Language Models through Domino LLM Gateway 2.0, the centrally governed gateway extension an administrator installs in front of external providers and Domino-hosted models. Covers finding and detecting the gateway, authentication with the in-run access token (fresh per call) or a Personal Access Token outside Domino, model aliases and model discovery, the OpenAI-compatible and Anthropic-compatible paths from the OpenAI and Anthropic SDKs, Pydantic AI, LangChain and plain HTTP, streaming, attribution headers, refused calls (login redirect, grants, budgets, guardrails), pointing a coding assistant such as Claude Code or Codex at the gateway, and what the administrator configures. Use when a user mentions LLM Gateway, Gateway 2.0, a gateway alias or URL, asks how to call a model through the Domino gateway, how to run a coding assistant through it, why a gateway call was refused, or what the gateway administrator must set up.
| File | Kind | Size |
|---|---|---|
| ADMIN-OVERVIEW.md | markdown | 2.1 KB |
| CALLING-MODELS.md | markdown | 8.6 KB |
| CODING-ASSISTANTS.md | markdown | 10.4 KB |
| SKILL.md | skill md | 7.7 KB |
ADMIN-OVERVIEW.md
View on GitHub# LLM Gateway 2.0: what the administrator configures
Docs: https://docs.domino.ai/cloud/platform-capabilities/features/llms/llm-gateway.
Everything below happens in the gateway's own UI, by the gateway's owner or an administrator; a developer does not need any of it to call a model, only to understand why a call is refused.
- **Providers and aliases.** The administrator connects upstream providers (external LLM vendors and Domino-hosted model endpoints) and publishes **model aliases**, possibly under a prefix such as `domino/`. An alias is the name callers pass as `model`; which vendor model sits behind it is the administrator's choice and can change without callers changing code. Callers see only the aliases they are granted (`GET <base>/v1/models`), which can be far fewer than the catalog.
- **Access and budgets.** Access is granted per alias to users, organizations or everyone; a grant can carry a budget for a period, and an alias can carry an overall budget. A call to an alias the caller has no grant for, or whose budget is spent, is refused with a message that says so; the caller asks the administrator for access or budget.
- **Guardrails.** The administrator can screen prompts and responses at call time with rules whose outcome is to deny, redact or log. A denied call returns an error naming the rule; redaction changes the content. This is separate from scoring finished traces with evaluators (`domino-genai-tracing` skill).
- **Usage and audit.** The gateway records every request (user, Project, alias, tokens, latency, estimated cost) and every configuration change; the administrator reads them in the gateway UI and can attribute spend by the tags callers send (`X-LLM-Tag-<key>` headers).
- **Tokens for use outside Domino.** The administrator can issue gateway tokens (prefix `dgw_`) with an expiry for scripts, CI and laptops. Which bearer works from where: [SKILL.md](./SKILL.md#authentication); inside Domino nobody needs a gateway token.
Installing, updating and operating the extension are not covered by this skill; point administrators to the material Domino provides with the gateway.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |