{"owner":"andrewyng","repo":"aisuite","hasSkills":true,"totalSkillsCount":4,"totalTokensCount":1200,"categories":["gemini-rule","root-instruction"],"hasMcp":false,"mcpConfig":null,"found":["guides/gemini.md","guides/mistral.md","guides/openai.md","guides/deepseek.md"],"skills":{"guides/gemini.md":"# Gemini (Gemini Developer API)\n\nThis provider talks to Google's [Gemini Developer API](https://ai.google.dev/) using a plain API key — no Google Cloud project, service account, or GCS bucket required. (For Vertex AI with GCP project auth, use the [Google](google.md) provider instead.)\n\n## Get an API key\n\nVisit [Google AI Studio](https://aistudio.google.com/apikey), sign in with a Google account, and click \"Create API key.\"\n\n## Set the API key in an environment variable\n\n```shell\nexport GEMINI_API_KEY=\"your-gemini-api-key\"\n```\n\n(`GOOGLE_API_KEY`, the SDK's own convention, works as a fallback.)\n\n## Install the SDK and use the provider\n\n```shell\npip install 'aisuite[gemini]'\n```\n\n```python\nimport aisuite as ai\n\nclient = ai.Client()\n\nresponse = client.chat.completions.create(\n    model=\"gemini:gemini-2.5-flash\",\n    messages=[{\"role\": \"user\", \"content\": \"Tell me a joke.\"}],\n)\nprint(response.choices[0].message.content)\n```\n\nTool calling, streaming (`stream=True`), and image inputs are supported. Images must be base64 data URLs (`data:image/png;base64,...`) in OpenAI-style `image_url` content parts — the Gemini API does not fetch plain http(s) URLs.\n","guides/mistral.md":"# Mistral\n\nTo use Mistral with `aisuite`, you’ll need a [Mistral account](https://console.mistral.ai/). \n\nAfter logging in, go to [Workspace billing](https://console.mistral.ai/billing) and choose a plan\n- **Experiment** *(Free, 1 request per second); or*\n- **Scale** *(Pay per use).*\n\nVisit the [API Keys](https://console.mistral.ai/api-keys/) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport MISTRAL=\"your-mistralai-api-key\"\n```\n## Create a Chat Completion\n\nInstall the `mistralai` Python client:\n\nExample with pip:\n```shell\npip install mistralai\n```\n\nExample with poetry:\n```shell\npoetry add mistralai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"mistral\"\nmodel_id = \"mistral-large-latest\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in Montréal?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n","guides/openai.md":"# OpenAI\n\nTo use OpenAI with `aisuite`, you’ll need an [OpenAI account](https://platform.openai.com/). After logging in, go to the [API Keys](https://platform.openai.com/account/api-keys) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport OPENAI_API_KEY=\"your-openai-api-key\"\n```\n\n## Create a Chat Completion\n\nInstall the `openai` Python client:\n\nExample with pip:\n```shell\npip install openai\n```\n\nExample with poetry:\n```shell\npoetry add openai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"openai\"\nmodel_id = \"gpt-4-turbo\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in San Francisco?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n","guides/deepseek.md":"# DeepSeek\n\nTo use DeepSeek with `aisuite`, you’ll need an [DeepSeek account](https://platform.deepseek.com). After logging in, go to the [API Keys](https://platform.deepseek.com/api_keys) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport DEEPSEEK_API_KEY=\"your-deepseek-api-key\"\n```\n\n## Create a Chat Completion\n\n(Note: The DeepSeek uses an API format consistent with OpenAI, hence why we need to install OpenAI, there is no DeepSeek Library at least not for now)\n\nInstall the `openai` Python client:\n\nExample with pip:\n```shell\npip install openai\n```\n\nExample with poetry:\n```shell\npoetry add openai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"deepseek\"\nmodel_id = \"deepseek-chat\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in San Francisco?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n"},"files":{"guides/gemini.md":"# Gemini (Gemini Developer API)\n\nThis provider talks to Google's [Gemini Developer API](https://ai.google.dev/) using a plain API key — no Google Cloud project, service account, or GCS bucket required. (For Vertex AI with GCP project auth, use the [Google](google.md) provider instead.)\n\n## Get an API key\n\nVisit [Google AI Studio](https://aistudio.google.com/apikey), sign in with a Google account, and click \"Create API key.\"\n\n## Set the API key in an environment variable\n\n```shell\nexport GEMINI_API_KEY=\"your-gemini-api-key\"\n```\n\n(`GOOGLE_API_KEY`, the SDK's own convention, works as a fallback.)\n\n## Install the SDK and use the provider\n\n```shell\npip install 'aisuite[gemini]'\n```\n\n```python\nimport aisuite as ai\n\nclient = ai.Client()\n\nresponse = client.chat.completions.create(\n    model=\"gemini:gemini-2.5-flash\",\n    messages=[{\"role\": \"user\", \"content\": \"Tell me a joke.\"}],\n)\nprint(response.choices[0].message.content)\n```\n\nTool calling, streaming (`stream=True`), and image inputs are supported. Images must be base64 data URLs (`data:image/png;base64,...`) in OpenAI-style `image_url` content parts — the Gemini API does not fetch plain http(s) URLs.\n","guides/mistral.md":"# Mistral\n\nTo use Mistral with `aisuite`, you’ll need a [Mistral account](https://console.mistral.ai/). \n\nAfter logging in, go to [Workspace billing](https://console.mistral.ai/billing) and choose a plan\n- **Experiment** *(Free, 1 request per second); or*\n- **Scale** *(Pay per use).*\n\nVisit the [API Keys](https://console.mistral.ai/api-keys/) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport MISTRAL=\"your-mistralai-api-key\"\n```\n## Create a Chat Completion\n\nInstall the `mistralai` Python client:\n\nExample with pip:\n```shell\npip install mistralai\n```\n\nExample with poetry:\n```shell\npoetry add mistralai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"mistral\"\nmodel_id = \"mistral-large-latest\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in Montréal?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n","guides/openai.md":"# OpenAI\n\nTo use OpenAI with `aisuite`, you’ll need an [OpenAI account](https://platform.openai.com/). After logging in, go to the [API Keys](https://platform.openai.com/account/api-keys) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport OPENAI_API_KEY=\"your-openai-api-key\"\n```\n\n## Create a Chat Completion\n\nInstall the `openai` Python client:\n\nExample with pip:\n```shell\npip install openai\n```\n\nExample with poetry:\n```shell\npoetry add openai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"openai\"\nmodel_id = \"gpt-4-turbo\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in San Francisco?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n","guides/deepseek.md":"# DeepSeek\n\nTo use DeepSeek with `aisuite`, you’ll need an [DeepSeek account](https://platform.deepseek.com). After logging in, go to the [API Keys](https://platform.deepseek.com/api_keys) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport DEEPSEEK_API_KEY=\"your-deepseek-api-key\"\n```\n\n## Create a Chat Completion\n\n(Note: The DeepSeek uses an API format consistent with OpenAI, hence why we need to install OpenAI, there is no DeepSeek Library at least not for now)\n\nInstall the `openai` Python client:\n\nExample with pip:\n```shell\npip install openai\n```\n\nExample with poetry:\n```shell\npoetry add openai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"deepseek\"\nmodel_id = \"deepseek-chat\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in San Francisco?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n"},"items":[{"name":"deepseek.md","path":"guides/deepseek.md","rawUrl":"https://raw.githubusercontent.com/andrewyng/aisuite/HEAD/guides/deepseek.md","title":"guides - DeepSeek System & Reasoning Rules","category":"root-instruction","format":"markdown","content":"# DeepSeek\n\nTo use DeepSeek with `aisuite`, you’ll need an [DeepSeek account](https://platform.deepseek.com). After logging in, go to the [API Keys](https://platform.deepseek.com/api_keys) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport DEEPSEEK_API_KEY=\"your-deepseek-api-key\"\n```\n\n## Create a Chat Completion\n\n(Note: The DeepSeek uses an API format consistent with OpenAI, hence why we need to install OpenAI, there is no DeepSeek Library at least not for now)\n\nInstall the `openai` Python client:\n\nExample with pip:\n```shell\npip install openai\n```\n\nExample with poetry:\n```shell\npoetry add openai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"deepseek\"\nmodel_id = \"deepseek-chat\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in San Francisco?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n","isInternal":false,"tokens":307,"sizeBytes":1232},{"name":"mistral.md","path":"guides/mistral.md","rawUrl":"https://raw.githubusercontent.com/andrewyng/aisuite/HEAD/guides/mistral.md","title":"guides - Mistral Agent & Codestral Directives","category":"root-instruction","format":"markdown","content":"# Mistral\n\nTo use Mistral with `aisuite`, you’ll need a [Mistral account](https://console.mistral.ai/). \n\nAfter logging in, go to [Workspace billing](https://console.mistral.ai/billing) and choose a plan\n- **Experiment** *(Free, 1 request per second); or*\n- **Scale** *(Pay per use).*\n\nVisit the [API Keys](https://console.mistral.ai/api-keys/) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport MISTRAL=\"your-mistralai-api-key\"\n```\n## Create a Chat Completion\n\nInstall the `mistralai` Python client:\n\nExample with pip:\n```shell\npip install mistralai\n```\n\nExample with poetry:\n```shell\npoetry add mistralai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"mistral\"\nmodel_id = \"mistral-large-latest\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in Montréal?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n","isInternal":false,"tokens":325,"sizeBytes":1238},{"name":"openai.md","path":"guides/openai.md","rawUrl":"https://raw.githubusercontent.com/andrewyng/aisuite/HEAD/guides/openai.md","title":"guides - OpenAI Codex & System Instructions","category":"root-instruction","format":"markdown","content":"# OpenAI\n\nTo use OpenAI with `aisuite`, you’ll need an [OpenAI account](https://platform.openai.com/). After logging in, go to the [API Keys](https://platform.openai.com/account/api-keys) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport OPENAI_API_KEY=\"your-openai-api-key\"\n```\n\n## Create a Chat Completion\n\nInstall the `openai` Python client:\n\nExample with pip:\n```shell\npip install openai\n```\n\nExample with poetry:\n```shell\npoetry add openai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"openai\"\nmodel_id = \"gpt-4-turbo\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in San Francisco?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n","isInternal":false,"tokens":275,"sizeBytes":1071},{"name":"gemini.md","path":"guides/gemini.md","rawUrl":"https://raw.githubusercontent.com/andrewyng/aisuite/HEAD/guides/gemini.md","title":"guides - Gemini Agent & Antigravity Instructions","category":"gemini-rule","format":"markdown","content":"# Gemini (Gemini Developer API)\n\nThis provider talks to Google's [Gemini Developer API](https://ai.google.dev/) using a plain API key — no Google Cloud project, service account, or GCS bucket required. (For Vertex AI with GCP project auth, use the [Google](google.md) provider instead.)\n\n## Get an API key\n\nVisit [Google AI Studio](https://aistudio.google.com/apikey), sign in with a Google account, and click \"Create API key.\"\n\n## Set the API key in an environment variable\n\n```shell\nexport GEMINI_API_KEY=\"your-gemini-api-key\"\n```\n\n(`GOOGLE_API_KEY`, the SDK's own convention, works as a fallback.)\n\n## Install the SDK and use the provider\n\n```shell\npip install 'aisuite[gemini]'\n```\n\n```python\nimport aisuite as ai\n\nclient = ai.Client()\n\nresponse = client.chat.completions.create(\n    model=\"gemini:gemini-2.5-flash\",\n    messages=[{\"role\": \"user\", \"content\": \"Tell me a joke.\"}],\n)\nprint(response.choices[0].message.content)\n```\n\nTool calling, streaming (`stream=True`), and image inputs are supported. Images must be base64 data URLs (`data:image/png;base64,...`) in OpenAI-style `image_url` content parts — the Gemini API does not fetch plain http(s) URLs.\n","isInternal":false,"tokens":293,"sizeBytes":1167}],"systemPromptSnippet":"<yakaai_skills repo=\"andrewyng/aisuite\">\n<!-- File: guides/deepseek.md (Tokens: ~307 | Category: root-instruction) -->\n# DeepSeek\n\nTo use DeepSeek with `aisuite`, you’ll need an [DeepSeek account](https://platform.deepseek.com). After logging in, go to the [API Keys](https://platform.deepseek.com/api_keys) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport DEEPSEEK_API_KEY=\"your-deepseek-api-key\"\n```\n\n## Create a Chat Completion\n\n(Note: The DeepSeek uses an API format consistent with OpenAI, hence why we need to install OpenAI, there is no DeepSeek Library at least not for now)\n\nInstall the `openai` Python client:\n\nExample with pip:\n```shell\npip install openai\n```\n\nExample with poetry:\n```shell\npoetry add openai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"deepseek\"\nmodel_id = \"deepseek-chat\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in San Francisco?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n\n<!-- File: guides/mistral.md (Tokens: ~325 | Category: root-instruction) -->\n# Mistral\n\nTo use Mistral with `aisuite`, you’ll need a [Mistral account](https://console.mistral.ai/). \n\nAfter logging in, go to [Workspace billing](https://console.mistral.ai/billing) and choose a plan\n- **Experiment** *(Free, 1 request per second); or*\n- **Scale** *(Pay per use).*\n\nVisit the [API Keys](https://console.mistral.ai/api-keys/) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport MISTRAL=\"your-mistralai-api-key\"\n```\n## Create a Chat Completion\n\nInstall the `mistralai` Python client:\n\nExample with pip:\n```shell\npip install mistralai\n```\n\nExample with poetry:\n```shell\npoetry add mistralai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"mistral\"\nmodel_id = \"mistral-large-latest\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in Montréal?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n\n<!-- File: guides/openai.md (Tokens: ~275 | Category: root-instruction) -->\n# OpenAI\n\nTo use OpenAI with `aisuite`, you’ll need an [OpenAI account](https://platform.openai.com/). After logging in, go to the [API Keys](https://platform.openai.com/account/api-keys) section in your account settings and generate a new key. Once you have your key, add it to your environment as follows:\n\n```shell\nexport OPENAI_API_KEY=\"your-openai-api-key\"\n```\n\n## Create a Chat Completion\n\nInstall the `openai` Python client:\n\nExample with pip:\n```shell\npip install openai\n```\n\nExample with poetry:\n```shell\npoetry add openai\n```\n\nIn your code:\n```python\nimport aisuite as ai\nclient = ai.Client()\n\nprovider = \"openai\"\nmodel_id = \"gpt-4-turbo\"\n\nmessages = [\n    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n    {\"role\": \"user\", \"content\": \"What’s the weather like in San Francisco?\"},\n]\n\nresponse = client.chat.completions.create(\n    model=f\"{provider}:{model_id}\",\n    messages=messages,\n)\n\nprint(response.choices[0].message.content)\n```\n\nHappy coding! If you’d like to contribute, please read our [Contributing Guide](../CONTRIBUTING.md).\n\n<!-- File: guides/gemini.md (Tokens: ~293 | Category: gemini-rule) -->\n# Gemini (Gemini Developer API)\n\nThis provider talks to Google's [Gemini Developer API](https://ai.google.dev/) using a plain API key — no Google Cloud project, service account, or GCS bucket required. (For Vertex AI with GCP project auth, use the [Google](google.md) provider instead.)\n\n## Get an API key\n\nVisit [Google AI Studio](https://aistudio.google.com/apikey), sign in with a Google account, and click \"Create API key.\"\n\n## Set the API key in an environment variable\n\n```shell\nexport GEMINI_API_KEY=\"your-gemini-api-key\"\n```\n\n(`GOOGLE_API_KEY`, the SDK's own convention, works as a fallback.)\n\n## Install the SDK and use the provider\n\n```shell\npip install 'aisuite[gemini]'\n```\n\n```python\nimport aisuite as ai\n\nclient = ai.Client()\n\nresponse = client.chat.completions.create(\n    model=\"gemini:gemini-2.5-flash\",\n    messages=[{\"role\": \"user\", \"content\": \"Tell me a joke.\"}],\n)\nprint(response.choices[0].message.content)\n```\n\nTool calling, streaming (`stream=True`), and image inputs are supported. Images must be base64 data URLs (`data:image/png;base64,...`) in OpenAI-style `image_url` content parts — the Gemini API does not fetch plain http(s) URLs.\n</yakaai_skills>"}