arcee-ai

Imported from aijadugar/skyboy (skills/model-providers/arcee-ai/SKILL.md). Install upstream with npx skills add aijadugar/skyboy --skill arcee-ai. Copyright stays with the author (MIT).

Command

Install this skill with:

skyboy add arcee-ai

Or preview it first with skyboy info arcee-ai.

Arcee AI

Use this skill when integrating with Arcee AI as a model provider. Covers API setup, authentication, model selection, and best practices for production use.

When it applies

• Setting up Arcee AI API credentials and SDK
• Choosing the right model from Arcee AI's offerings
• Handling rate limits, retries, and error responses
• Optimizing cost and latency with Arcee AI
• Streaming responses and tool use with Arcee AI

Quick start

# Install the SDK (example — adapt to the provider's actual package)
npm install @arcee-ai/sdk

import { ArceeAIClient } from "@arcee-ai/sdk";

const client = new ArceeAIClient({
  apiKey: process.env.ARCEE_AI_API_KEY,
});

Model selection

Arcee AI offers models across capability tiers. Choose based on your use case:

Best practices

• Rate limits: Respect Arcee AI's published RPM/TPM limits; implement exponential backoff.
• Streaming: Use streaming for interactive UIs to reduce time-to-first-token.
• Cost control: Cache responses where possible; use shorter prompts with system instructions.
• Error handling: Map Arcee AI's error codes to user-friendly messages; never leak API keys client-side.

References

• Arcee AI official documentation
• Arcee AI API reference

Arcee AI Skills & Plugins

Arcee's public skills and plugin ecosystem (skills/ and plugins/ in this folder):

Skills

Plugins