Oneworks model services
Skill oneworks-ai/app/packages/plugins/cli-skills/skills/oneworks-model-services
One Works application monorepo for AI agents, plugins, desktop, web, CLI, and Relay.
npx -y skills add oneworks-ai/app --skill oneworks-model-servicesAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 12 stars12 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Configure OneWorks model services using built-in provider defaults, management portals, API base URLs, model lists, balance/status capabilities, and safe config-source writeback.
SKILL.md
13.5 KB, as published. Nobody here has run it
Use this skill when a user wants to add, update, or troubleshoot OneWorks modelServices.
Writeback Rules
- Put account-level API keys and secrets in global config by default.
- Use project or user config only when the user explicitly asks for an override.
- Do not copy values from merged config back into a source file unless you know the target source.
- Never show API keys, secrets, tokens, or passwords in the final answer.
- If the selected provider is known, prefer built-in defaults for
provider,homepageUrl,apiBaseUrl, models, icons, and management capabilities. Only write override fields when the user asks for an override.
Config Shape
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 reuseapiKeywhen supported.
Coding Plan and Token Plan Rules
Here "Coding Plan" means a provider billing/quota product, not OneWorks or agent Plan Mode.
- Do not mix plan keys and ordinary pay-as-you-go API keys. A Coding Plan or Token Plan normally needs its dedicated key and dedicated base URL.
- Prefer built-in plan provider ids instead of overriding ordinary API providers:
kimi-code: Kimi Code, OpenAIhttps://api.kimi.com/coding/v1, Anthropichttps://api.kimi.com/coding/, modelkimi-for-coding, Kimi Code API key from the Kimi Code console. Models and quota can be queried through the dedicated/coding/v1API; do not use Moonshot/Kimi Platform PAYG balance endpoints for it.minimax-token-plan: MiniMax Token Plan, OpenAIhttps://api.minimax.io/v1, Anthropichttps://api.minimax.io/anthropic, China Anthropichttps://api.minimaxi.com/anthropic, modelMiniMax-M3, Subscription Key.qwen-coding-plan: Alibaba Cloud Model Studio Coding Plan, OpenAIhttps://coding.dashscope.aliyuncs.com/v1, Anthropichttps://coding.dashscope.aliyuncs.com/apps/anthropic, international OpenAIhttps://coding-intl.dashscope.aliyuncs.com/v1, international Anthropichttps://coding-intl.dashscope.aliyuncs.com/apps/anthropic, plan key commonlysk-sp-*.zhipu-coding-plan: Zhipu GLM Coding Plan, OpenAIhttps://open.bigmodel.cn/api/coding/paas/v4, Anthropichttps://open.bigmodel.cn/api/anthropic, modelGLM-5.2.tencent-tokenhub-coding-plan: Tencent Cloud TokenHub Coding Plan, OpenAIhttps://api.lkeap.cloud.tencent.com/coding/v3, Anthropichttps://api.lkeap.cloud.tencent.com/coding/anthropic.volcengine-ark-coding-plan: Volcengine Ark Coding Plan, OpenAIhttps://ark.cn-beijing.volces.com/api/coding/v3, Anthropichttps://ark.cn-beijing.volces.com/api/coding.baidu-qianfan-coding-plan: Baidu Qianfan Coding Plan, OpenAIhttps://qianfan.baidubce.com/v2/coding, Anthropichttps://qianfan.baidubce.com/anthropic/coding.
- OpenAI Codex and Anthropic Claude Code subscriptions are product-login entitlements, not model service API keys. ChatGPT/Codex and
claude logincan use subscription quotas, while OpenAI/Anthropic API keys still bill as ordinary API usage. - Many plan quotas are request-window based, such as 5-hour, weekly, or monthly request limits. Do not describe them as ordinary token quotas unless the provider calls the product a Token Plan.
- Many providers restrict plans to interactive coding tools and disallow backend batch jobs, automation scripts, or generic API load. Preserve
billing.allowedUse: coding_tools_onlyand show the restriction to the user. - Plan model lists are catalog defaults and may not be available from
/v1/models. Leavemodelsempty to use the built-in catalog, or writemodelsonly when the user explicitly wants an allowlist override. - DeepSeek is currently a pay-as-you-go API provider with coding-agent integrations, not a dedicated Coding Plan provider in this registry.
Built-In Providers
Official providers:
openai: OpenAI,https://api.openai.com/v1; portalhttps://platform.openai.com; API keyshttps://platform.openai.com/api-keys; docshttps://platform.openai.com/docs; statushttps://status.openai.com; model listing via API; balance unsupported.anthropic: Anthropic,https://api.anthropic.com/v1; portalhttps://console.anthropic.com; API keyshttps://console.anthropic.com/settings/keys; billinghttps://console.anthropic.com/settings/billing; docshttps://docs.anthropic.com; statushttps://status.claude.com; model listing manual.- Claude Code should use Anthropic directly with
ANTHROPIC_API_KEY; OneWorks normalizeshttps://api.anthropic.com/v1tohttps://api.anthropic.comfor the native CLI.
- Claude Code should use Anthropic directly with
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 arekimi-k2.7-code,kimi-k2.6,kimi-k2.5,kimi-k2-0905-preview,kimi-k2. Portalhttps://platform.kimi.com; API keyshttps://platform.kimi.com/console/api-keys; billinghttps://platform.kimi.com/console/account; docshttps://platform.kimi.com/docs; statushttps://status.moonshot.cn.- Claude Code should use Kimi's official Anthropic API adapter at
https://api.moonshot.cn/anthropicinstead of routing through Claude Code Router's OpenAI conversion path.
- Claude Code should use Kimi's official Anthropic API adapter at
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 arekimi-k2.7-code,kimi-k2.6,kimi-k2.5,kimi-k2-0905-preview,kimi-k2. Portalhttps://platform.kimi.ai; API keyshttps://platform.kimi.ai/console/api-keys; billinghttps://platform.kimi.ai/console/account; docshttps://platform.kimi.ai/docs; statushttps://status.moonshot.cn.- Claude Code should use Kimi's official Anthropic API adapter at
https://api.moonshot.ai/anthropicinstead of routing through Claude Code Router's OpenAI conversion path.
- Claude Code should use Kimi's official Anthropic API adapter at
deepseek: DeepSeek,https://api.deepseek.com; default modelsdeepseek-v4-flash,deepseek-v4-pro, with legacy aliasesdeepseek-chat,deepseek-reasonerkept for compatibility until their documented 2026-07-24 deprecation; portalhttps://platform.deepseek.com; API keyshttps://platform.deepseek.com/api_keys; billinghttps://platform.deepseek.com/usage; docshttps://api-docs.deepseek.com; statushttps://status.deepseek.com; model and balance APIs are supported.- Claude Code should use DeepSeek's official Anthropic API adapter at
https://api.deepseek.com/anthropicinstead of routing through Claude Code Router's OpenAI conversion path.
- Claude Code should use DeepSeek's official Anthropic API adapter at
minimax: MiniMax,https://api.minimax.io/v1; default modelMiniMax-M3; portalhttps://platform.minimaxi.com; API keyshttps://platform.minimaxi.com/user-center/basic-information/interface-key; billinghttps://platform.minimaxi.com/user-center/basic-information/interface-balance; docshttps://platform.minimaxi.com/document; statushttps://status.minimax.io; model API supported, balance manual.- Claude Code should use MiniMax's official Anthropic API adapter at
https://api.minimax.io/anthropicfor international accounts orhttps://api.minimaxi.com/anthropicfor China accounts. OneWorks preserves the configured host and switches the path to/anthropic.
- Claude Code should use MiniMax's official Anthropic API adapter at
minimax-token-plan: MiniMax Token Plan; use Subscription Key, not a pay-as-you-go API key. Leavemodelsempty forMiniMax-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 modelsqwen3.7-max,qwen3.7-plus,qwen3.6-flash,qwen3.5-plus,qwen3-coder-next,qwen3-coder-plus; portalhttps://bailian.console.aliyun.com; API keyshttps://bailian.console.aliyun.com/?tab=api#/api-key; billinghttps://usercenter2.aliyun.com/finance; docshttps://help.aliyun.com/zh/model-studio; model list is static.- Claude Code should use Bailian's official Anthropic-compatible endpoints, for example pay-as-you-go
https://dashscope.aliyuncs.com/apps/anthropic, Coding Planhttps://coding.dashscope.aliyuncs.com/apps/anthropic, Token Planhttps://token-plan.cn-beijing.maas.aliyuncs.com/apps/anthropic, or the regional workspace host ending in/apps/anthropic.
- Claude Code should use Bailian's official Anthropic-compatible endpoints, for example pay-as-you-go
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 modelsglm-5.2[1m],glm-5.2,glm-4.7,glm-4.5-air; portalhttps://open.bigmodel.cn; API keyshttps://open.bigmodel.cn/usercenter/proj-mgmt/apikeys; billinghttps://open.bigmodel.cn/usercenter/resourcepool; docshttps://docs.bigmodel.cn; model list is static.- Claude Code should use Zhipu's official Anthropic-compatible endpoint at
https://open.bigmodel.cn/api/anthropic. Forglm-5.2[1m], keepCLAUDE_CODE_AUTO_COMPACT_WINDOW=1000000; useglm-4.5-airfor the Haiku slot when available.
- Claude Code should use Zhipu's official Anthropic-compatible endpoint at
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, andbaidu-qianfan-coding-planare dedicated coding subscription providers with static model catalogs and dedicated OpenAI/Anthropic-compatible endpoints.
Cloud providers:
azure-openai: Azure OpenAI through Azure AI Foundry; portalhttps://ai.azure.com; billing in Azure portal; docshttps://learn.microsoft.com/azure/ai-services/openai; statushttps://status.azure.com.google-gemini: Google Gemini OpenAI-compatible endpointhttps://generativelanguage.googleapis.com/v1beta/openai; portalhttps://aistudio.google.com; API keyshttps://aistudio.google.com/apikey; billinghttps://console.cloud.google.com/billing; docshttps://ai.google.dev/gemini-api/docs.aws-bedrock: AWS Bedrock; portalhttps://console.aws.amazon.com/bedrock; billinghttps://console.aws.amazon.com/billing; docshttps://docs.aws.amazon.com/bedrock; statushttps://health.aws.amazon.com/health/status.
Relay and gateway providers:
openrouter:https://openrouter.ai/api/v1; portalhttps://openrouter.ai; API keyshttps://openrouter.ai/settings/keys; billinghttps://openrouter.ai/settings/credits; docshttps://openrouter.ai/docs.- Claude Code should use OpenRouter's Anthropic Skin directly at
https://openrouter.ai/api, withANTHROPIC_AUTH_TOKENand emptyANTHROPIC_API_KEY.
- Claude Code should use OpenRouter's Anthropic Skin directly at
vercel-ai-gateway:https://ai-gateway.vercel.sh/v1; portalhttps://vercel.com/ai-gateway; billinghttps://vercel.com/dashboard/usage; docshttps://vercel.com/docs/ai-gateway.- Claude Code should use Vercel AI Gateway's Anthropic-compatible endpoint at
https://ai-gateway.vercel.sh, withANTHROPIC_AUTH_TOKENand emptyANTHROPIC_API_KEY.
- Claude Code should use Vercel AI Gateway's Anthropic-compatible endpoint at
requesty:https://router.requesty.ai/v1; portalhttps://requesty.ai; API keyshttps://app.requesty.ai/api-keys; billinghttps://app.requesty.ai/billing; docshttps://docs.requesty.ai.- Claude Code should use Requesty's Anthropic-compatible gateway at
https://router.requesty.ai, with model IDs such asprovider/model-nameorpolicy/name.
- Claude Code should use Requesty's Anthropic-compatible gateway at
portkey:https://api.portkey.ai/v1; portalhttps://portkey.ai; API keyshttps://app.portkey.ai/api-keys; billinghttps://app.portkey.ai/billing; docshttps://portkey.ai/docs.- Claude Code should use Portkey's unified endpoint at
https://api.portkey.ai. If a Portkey provider slug is required, configureextra.claudeCode.portkeyProvider, or set fullextra.claudeCode.anthropicCustomHeaders.
- Claude Code should use Portkey's unified endpoint at
litellm: LiteLLM gateway; portalhttps://www.litellm.ai; docshttps://docs.litellm.ai.micu: Micu relay; portal and docshttps://micu.hk.apiyi: APIYI relay; portal and docshttps://apiyi.com.yunwu: Yunwu relay; portal and docshttps://yunwu.ai.custom-openai-compatible: user-defined OpenAI-compatible endpoint; requires explicitapiBaseUrl, usually explicitmodels, and manual management links.
Workflow
- Confirm provider, region/site when relevant, API key, whether defaults should be used, and whether management helpers should be enabled.
- If the provider has a portal, guide the user to open the portal for login, billing/top-up, and secret/API-key creation.
- Write the minimal config: use
providerplusapiKey; omitapiBaseUrl,models,homepageUrl, and icon fields when built-in defaults are enough. - For custom OpenAI-compatible services, ask for
apiBaseUrl, supported model ids, homepage, and any billing/API-key URLs. - After saving, verify with model listing or a short chat request when credentials are available.