메뉴
Configure OneWorks model services using built-in provider defaults, management portals, API base URLs, model lists, balance/status capabilities, and safe config-source writeback.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
Control the current OneWorks session's internal browser tabs with semantic snapshots, reliable element refs, screenshots, waits, and page-safe concurrent workflows.
Drive native macOS apps through the Cua Driver tools. Use when a task asks to open, inspect, click, type, scroll, capture a final screenshot, or verify a real macOS app without stealing the user's focus.
Control a user-paired external browser through a Chrome extension, explicit typed targets, and semantic operations.
当用户想创建或改造 OneWorks plugin，实现界面入口、按钮、tab、launcher 搜索、server 命令、scoped API 或本地服务时，先理解目标效果；需求不明确时列出不确定点让用户确认，再转成 manifest、前端入口、server 入口和验证步骤。
Apply, review, and continuously evolve the OneWorks team's durable UI design standards. Use for any OneWorks task that changes or reviews layout, styling, theme, spacing, responsive behavior, component appearance, visual assets, or reference-image fidelity, and whenever user visual feedback may express a reusable team standard. Require a Visual Brief, conflict-aware project memory capture, a completed independent visual consistency session with real expected-behavior validation, and experience persistence before delivery; do not promote one-off pixel nudges into lasting rules.
快速说明 One Works CLI 的常用命令、配置命令、会话查看和基础技能选择方式。
| name | oneworks-model-services |
| description | Configure OneWorks model services using built-in provider defaults, management portals, API base URLs, model lists, balance/status capabilities, and safe config-source writeback. |
Use this skill when a user wants to add, update, or troubleshoot OneWorks modelServices.
provider, homepageUrl, apiBaseUrl, models, icons, and management capabilities. Only write override fields when the user asks for an override.Use modelServices.<serviceKey> with these common fields:
provider: built-in provider id.title: display name.description: short display description.apiKey: service API key or secret.apiBaseUrl: only needed for custom services or explicit overrides.models: only needed for custom services or explicit overrides.billing: optional metadata for billing kind, key kind, quota unit, quota windows, and allowed use.codingPlan: optional metadata for plan links, protocol base URLs, regions, default models, and restrictions.homepageUrl: only needed to override the provider portal.management.enabled: enables model/balance/status/secret helper actions when available.management.apiKey: optional separate management API key; otherwise reuse apiKey when supported.Here "Coding Plan" means a provider billing/quota product, not OneWorks or agent Plan Mode.
kimi-code: Kimi Code, OpenAI https://api.kimi.com/coding/v1, Anthropic https://api.kimi.com/coding/, model kimi-for-coding, Kimi Code API key from the Kimi Code console. Models and quota can be queried through the dedicated /coding/v1 API; do not use Moonshot/Kimi Platform PAYG balance endpoints for it.minimax-token-plan: MiniMax Token Plan, OpenAI https://api.minimax.io/v1, Anthropic https://api.minimax.io/anthropic, China Anthropic https://api.minimaxi.com/anthropic, model MiniMax-M3, Subscription Key.qwen-coding-plan: Alibaba Cloud Model Studio Coding Plan, OpenAI https://coding.dashscope.aliyuncs.com/v1, Anthropic https://coding.dashscope.aliyuncs.com/apps/anthropic, international OpenAI https://coding-intl.dashscope.aliyuncs.com/v1, international Anthropic https://coding-intl.dashscope.aliyuncs.com/apps/anthropic, plan key commonly sk-sp-*.zhipu-coding-plan: Zhipu GLM Coding Plan, OpenAI https://open.bigmodel.cn/api/coding/paas/v4, Anthropic https://open.bigmodel.cn/api/anthropic, model GLM-5.2.tencent-tokenhub-coding-plan: Tencent Cloud TokenHub Coding Plan, OpenAI https://api.lkeap.cloud.tencent.com/coding/v3, Anthropic https://api.lkeap.cloud.tencent.com/coding/anthropic.volcengine-ark-coding-plan: Volcengine Ark Coding Plan, OpenAI https://ark.cn-beijing.volces.com/api/coding/v3, Anthropic https://ark.cn-beijing.volces.com/api/coding.baidu-qianfan-coding-plan: Baidu Qianfan Coding Plan, OpenAI https://qianfan.baidubce.com/v2/coding, Anthropic https://qianfan.baidubce.com/anthropic/coding.claude login can use subscription quotas, while OpenAI/Anthropic API keys still bill as ordinary API usage.billing.allowedUse: coding_tools_only and show the restriction to the user./v1/models. Leave models empty to use the built-in catalog, or write models only when the user explicitly wants an allowlist override.Official providers:
openai: OpenAI, https://api.openai.com/v1; portal https://platform.openai.com; API keys https://platform.openai.com/api-keys; docs https://platform.openai.com/docs; status https://status.openai.com; model listing via API; balance unsupported.anthropic: Anthropic, https://api.anthropic.com/v1; portal https://console.anthropic.com; API keys https://console.anthropic.com/settings/keys; billing https://console.anthropic.com/settings/billing; docs https://docs.anthropic.com; status https://status.claude.com; model listing manual.
ANTHROPIC_API_KEY; OneWorks normalizes https://api.anthropic.com/v1 to https://api.anthropic.com for the native CLI.moonshot-cn: Moonshot/Kimi China, https://api.moonshot.cn/v1; model and balance APIs are supported, so prefer refreshing and saving the provider model list, then use the first saved model as the default candidate. Built-in fallback models are kimi-k2.7-code, kimi-k2.6, kimi-k2.5, kimi-k2-0905-preview, kimi-k2. Portal https://platform.kimi.com; API keys https://platform.kimi.com/console/api-keys; billing https://platform.kimi.com/console/account; docs https://platform.kimi.com/docs; status https://status.moonshot.cn.
https://api.moonshot.cn/anthropic instead of routing through Claude Code Router's OpenAI conversion path.moonshot-intl: Moonshot/Kimi International, https://api.moonshot.ai/v1; model and balance APIs are supported, so prefer refreshing and saving the provider model list, then use the first saved model as the default candidate. Built-in fallback models are kimi-k2.7-code, kimi-k2.6, kimi-k2.5, kimi-k2-0905-preview, kimi-k2. Portal https://platform.kimi.ai; API keys https://platform.kimi.ai/console/api-keys; billing https://platform.kimi.ai/console/account; docs https://platform.kimi.ai/docs; status https://status.moonshot.cn.
https://api.moonshot.ai/anthropic instead of routing through Claude Code Router's OpenAI conversion path.deepseek: DeepSeek, https://api.deepseek.com; default models deepseek-v4-flash, deepseek-v4-pro, with legacy aliases deepseek-chat, deepseek-reasoner retained only for existing configurations after their documented 2026-07-24 deprecation; portal https://platform.deepseek.com; API keys https://platform.deepseek.com/api_keys; billing https://platform.deepseek.com/usage; docs https://api-docs.deepseek.com; status https://status.deepseek.com; model and balance APIs are supported.
deepseek-v4-flash is compatible; DeepSeek documents deepseek-v4-pro support as planned for early August 2026.https://api.deepseek.com/anthropic instead of routing through Claude Code Router's OpenAI conversion path.minimax: MiniMax, https://api.minimax.io/v1; default model MiniMax-M3; portal https://platform.minimaxi.com; API keys https://platform.minimaxi.com/user-center/basic-information/interface-key; billing https://platform.minimaxi.com/user-center/basic-information/interface-balance; docs https://platform.minimaxi.com/document; status https://status.minimax.io; model API supported, balance manual.
https://api.minimax.io/anthropic for international accounts or https://api.minimaxi.com/anthropic for China accounts. OneWorks preserves the configured host and switches the path to /anthropic.minimax-token-plan: MiniMax Token Plan; use Subscription Key, not a pay-as-you-go API key. Leave models empty for MiniMax-M3; Claude Code should use the plan Anthropic endpoint from the registry.qwen: Alibaba Qwen / Bailian, https://dashscope.aliyuncs.com/compatible-mode/v1; default Claude Code compatible models qwen3.7-max, qwen3.7-plus, qwen3.6-flash, qwen3.5-plus, qwen3-coder-next, qwen3-coder-plus; portal https://bailian.console.aliyun.com; API keys https://bailian.console.aliyun.com/?tab=api#/api-key; billing https://usercenter2.aliyun.com/finance; docs https://help.aliyun.com/zh/model-studio; model list is static.
https://dashscope.aliyuncs.com/apps/anthropic, Coding Plan https://coding.dashscope.aliyuncs.com/apps/anthropic, Token Plan https://token-plan.cn-beijing.maas.aliyuncs.com/apps/anthropic, or the regional workspace host ending in /apps/anthropic.qwen-coding-plan: Alibaba Cloud Model Studio Coding Plan; use the Coding Plan key with the Coding Plan endpoint. Do not use the ordinary DashScope key/base URL for this service.zhipu: Zhipu GLM, https://open.bigmodel.cn/api/paas/v4; default Claude Code compatible models glm-5.2[1m], glm-5.2, glm-4.7, glm-4.5-air; portal https://open.bigmodel.cn; API keys https://open.bigmodel.cn/usercenter/proj-mgmt/apikeys; billing https://open.bigmodel.cn/usercenter/resourcepool; docs https://docs.bigmodel.cn; model list is static.
https://open.bigmodel.cn/api/anthropic. For glm-5.2[1m], keep CLAUDE_CODE_AUTO_COMPACT_WINDOW=1000000; use glm-4.5-air for the Haiku slot when available.zhipu-coding-plan: Zhipu GLM Coding Plan; use the coding endpoint and coding-plan key. Ordinary BigModel endpoints do not count against the plan.kimi-code, tencent-tokenhub-coding-plan, volcengine-ark-coding-plan, and baidu-qianfan-coding-plan are dedicated coding subscription providers with static model catalogs and dedicated OpenAI/Anthropic-compatible endpoints.Cloud providers:
azure-openai: Azure OpenAI through Azure AI Foundry; portal https://ai.azure.com; billing in Azure portal; docs https://learn.microsoft.com/azure/ai-services/openai; status https://status.azure.com.google-gemini: Google Gemini OpenAI-compatible endpoint https://generativelanguage.googleapis.com/v1beta/openai; portal https://aistudio.google.com; API keys https://aistudio.google.com/apikey; billing https://console.cloud.google.com/billing; docs https://ai.google.dev/gemini-api/docs.aws-bedrock: AWS Bedrock; portal https://console.aws.amazon.com/bedrock; billing https://console.aws.amazon.com/billing; docs https://docs.aws.amazon.com/bedrock; status https://health.aws.amazon.com/health/status.Relay and gateway providers:
openrouter: https://openrouter.ai/api/v1; portal https://openrouter.ai; API keys https://openrouter.ai/settings/keys; billing https://openrouter.ai/settings/credits; docs https://openrouter.ai/docs.
https://openrouter.ai/api, with ANTHROPIC_AUTH_TOKEN and empty ANTHROPIC_API_KEY.vercel-ai-gateway: https://ai-gateway.vercel.sh/v1; portal https://vercel.com/ai-gateway; billing https://vercel.com/dashboard/usage; docs https://vercel.com/docs/ai-gateway.
https://ai-gateway.vercel.sh, with ANTHROPIC_AUTH_TOKEN and empty ANTHROPIC_API_KEY.requesty: https://router.requesty.ai/v1; portal https://requesty.ai; API keys https://app.requesty.ai/api-keys; billing https://app.requesty.ai/billing; docs https://docs.requesty.ai.
https://router.requesty.ai, with model IDs such as provider/model-name or policy/name.portkey: https://api.portkey.ai/v1; portal https://portkey.ai; API keys https://app.portkey.ai/api-keys; billing https://app.portkey.ai/billing; docs https://portkey.ai/docs.
https://api.portkey.ai. If a Portkey provider slug is required, configure extra.claudeCode.portkeyProvider, or set full extra.claudeCode.anthropicCustomHeaders.litellm: LiteLLM gateway; portal https://www.litellm.ai; docs https://docs.litellm.ai.micu: Micu relay; portal and docs https://micu.hk.apiyi: APIYI relay; portal and docs https://apiyi.com.yunwu: Yunwu relay; portal and docs https://yunwu.ai.custom-openai-compatible: user-defined OpenAI-compatible endpoint; requires explicit apiBaseUrl, usually explicit models, and manual management links.provider plus apiKey; omit apiBaseUrl, models, homepageUrl, and icon fields when built-in defaults are enough.apiBaseUrl, supported model ids, homepage, and any billing/API-key URLs.