@f5-sales-demo/pi-ai 20.19.1 → 20.19.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +34 -45
  2. package/package.json +2 -2
package/README.md CHANGED
@@ -46,8 +46,7 @@ Unified LLM API with automatic model discovery, provider configuration, token an
46
46
 
47
47
  ## Supported Providers
48
48
 
49
- - **OpenAI**
50
- - **OpenAI Codex** (ChatGPT Plus/Pro subscription, requires OAuth, see below)
49
+ - **OpenAI** (usage-based Platform API access)
51
50
  - **Anthropic**
52
51
  - **Google**
53
52
  - **Vertex AI** (Gemini via Vertex AI)
@@ -907,35 +906,35 @@ const response = await complete(
907
906
 
908
907
  In Node.js environments, you can set environment variables to avoid passing API keys:
909
908
 
910
- | Provider | Environment Variable(s) |
911
- | -------------- | ---------------------------------------------------------------------------- |
912
- | OpenAI | `OPENAI_API_KEY` |
913
- | Anthropic | `ANTHROPIC_API_KEY` or `ANTHROPIC_OAUTH_TOKEN` (or `ANTHROPIC_FOUNDRY_API_KEY` when `CLAUDE_CODE_USE_FOUNDRY=true`) |
914
- | Google | `GEMINI_API_KEY` |
915
- | Vertex AI | `GOOGLE_CLOUD_PROJECT` (or `GCLOUD_PROJECT`) + `GOOGLE_CLOUD_LOCATION` + ADC |
916
- | Mistral | `MISTRAL_API_KEY` |
917
- | Groq | `GROQ_API_KEY` |
918
- | Cerebras | `CEREBRAS_API_KEY` |
919
- | Together | `TOGETHER_API_KEY` |
920
- | Qianfan | `QIANFAN_API_KEY` |
921
- | Hugging Face | `HUGGINGFACE_HUB_TOKEN` or `HF_TOKEN` |
922
- | Synthetic | `SYNTHETIC_API_KEY` |
923
- | NVIDIA | `NVIDIA_API_KEY` |
924
- | NanoGPT | `NANO_GPT_API_KEY` |
925
- | Venice | `VENICE_API_KEY` |
926
- | Moonshot | `MOONSHOT_API_KEY` |
927
- | xAI | `XAI_API_KEY` |
928
- | OpenRouter | `OPENROUTER_API_KEY` |
929
- | LiteLLM | `LITELLM_API_KEY` |
930
- | Ollama | `OLLAMA_API_KEY` (optional for local deployments) |
931
- | Qwen Portal | `QWEN_OAUTH_TOKEN` or `QWEN_PORTAL_API_KEY` |
932
- | zAI | `ZAI_API_KEY` |
933
- | MiniMax Code | `MINIMAX_CODE_API_KEY` (international) or `MINIMAX_CODE_CN_API_KEY` (China) |
934
- | Xiaomi MiMo | `XIAOMI_API_KEY` |
935
- | ZenMux | `ZENMUX_API_KEY` |
936
- | vLLM | `VLLM_API_KEY` |
937
- | Cloudflare AI Gateway | `CLOUDFLARE_AI_GATEWAY_API_KEY` |
938
- | GitHub Copilot | `COPILOT_GITHUB_TOKEN` or `GH_TOKEN` or `GITHUB_TOKEN` |
909
+ | Provider | Environment Variable(s) |
910
+ | --------------------- | ------------------------------------------------------------------------------------------------------------------- |
911
+ | OpenAI | `OPENAI_API_KEY` |
912
+ | Anthropic | `ANTHROPIC_API_KEY` or `ANTHROPIC_OAUTH_TOKEN` (or `ANTHROPIC_FOUNDRY_API_KEY` when `CLAUDE_CODE_USE_FOUNDRY=true`) |
913
+ | Google | `GEMINI_API_KEY` |
914
+ | Vertex AI | `GOOGLE_CLOUD_PROJECT` (or `GCLOUD_PROJECT`) + `GOOGLE_CLOUD_LOCATION` + ADC |
915
+ | Mistral | `MISTRAL_API_KEY` |
916
+ | Groq | `GROQ_API_KEY` |
917
+ | Cerebras | `CEREBRAS_API_KEY` |
918
+ | Together | `TOGETHER_API_KEY` |
919
+ | Qianfan | `QIANFAN_API_KEY` |
920
+ | Hugging Face | `HUGGINGFACE_HUB_TOKEN` or `HF_TOKEN` |
921
+ | Synthetic | `SYNTHETIC_API_KEY` |
922
+ | NVIDIA | `NVIDIA_API_KEY` |
923
+ | NanoGPT | `NANO_GPT_API_KEY` |
924
+ | Venice | `VENICE_API_KEY` |
925
+ | Moonshot | `MOONSHOT_API_KEY` |
926
+ | xAI | `XAI_API_KEY` |
927
+ | OpenRouter | `OPENROUTER_API_KEY` |
928
+ | LiteLLM | `LITELLM_API_KEY` |
929
+ | Ollama | `OLLAMA_API_KEY` (optional for local deployments) |
930
+ | Qwen Portal | `QWEN_OAUTH_TOKEN` or `QWEN_PORTAL_API_KEY` |
931
+ | zAI | `ZAI_API_KEY` |
932
+ | MiniMax Code | `MINIMAX_CODE_API_KEY` (international) or `MINIMAX_CODE_CN_API_KEY` (China) |
933
+ | Xiaomi MiMo | `XIAOMI_API_KEY` |
934
+ | ZenMux | `ZENMUX_API_KEY` |
935
+ | vLLM | `VLLM_API_KEY` |
936
+ | Cloudflare AI Gateway | `CLOUDFLARE_AI_GATEWAY_API_KEY` |
937
+ | GitHub Copilot | `COPILOT_GITHUB_TOKEN` or `GH_TOKEN` or `GITHUB_TOKEN` |
939
938
 
940
939
  For Cloudflare AI Gateway models, use provider base URL format
941
940
  `https://gateway.ai.cloudflare.com/v1/<account>/<gateway>/anthropic`.
@@ -988,7 +987,6 @@ const key = getEnvApiKey("openai"); // checks OPENAI_API_KEY
988
987
  Several providers support OAuth authentication (some also support static API keys):
989
988
 
990
989
  - **Anthropic** (Claude Pro/Max subscription)
991
- - **OpenAI Codex** (ChatGPT Plus/Pro subscription, access to GPT-5.x Codex models)
992
990
  - **GitHub Copilot** (Copilot subscription)
993
991
  - **Google Gemini CLI** (Gemini 2.0/2.5 via Google Cloud Code Assist; free tier or paid subscription)
994
992
  - **Antigravity** (Free Gemini 3, Claude, GPT-OSS via Google Cloud)
@@ -1047,7 +1045,8 @@ bunx @f5-sales-demo/pi-ai list # list available providers
1047
1045
 
1048
1046
  Credentials are saved to `agent.db` in the agent directory. `/login qianfan` opens the Qianfan console and stores the pasted API key.
1049
1047
 
1050
- `login` supports OAuth providers (Anthropic, OpenAI Codex, GitHub Copilot, Gemini CLI, Antigravity) and API-key onboarding flows.
1048
+ `login` supports OAuth providers (Anthropic, GitHub Copilot, Gemini CLI, Antigravity) and API-key onboarding flows.
1049
+ OpenAI uses usage-based Platform API access through `OPENAI_API_KEY`. For ChatGPT subscription access, use the official Codex CLI and run `codex login`.
1051
1050
 
1052
1051
  For the current OpenAI-compatible integrations, API-key onboarding covers Together, Moonshot, Qianfan, NVIDIA, NanoGPT, Hugging Face, Venice, Xiaomi, vLLM, LiteLLM, Cloudflare AI Gateway, and Qwen Portal. Ollama is typically local and unauthenticated; set `OLLAMA_API_KEY` only when your Ollama deployment enforces bearer auth.
1053
1052
 
@@ -1059,7 +1058,6 @@ The library provides login and token refresh functions. Credential storage is th
1059
1058
  import {
1060
1059
  // Login functions (return credentials, do not store)
1061
1060
  loginAnthropic,
1062
- loginOpenAICodex,
1063
1061
  loginGitHubCopilot,
1064
1062
  loginGeminiCli,
1065
1063
  loginAntigravity,
@@ -1081,20 +1079,11 @@ import {
1081
1079
  getOAuthApiKey, // (provider, credentialsMap) => { newCredentials, apiKey } | null
1082
1080
 
1083
1081
  // Types
1084
- type OAuthProvider, // includes 'anthropic', 'openai-codex', 'github-copilot', 'google-gemini-cli', 'google-antigravity', 'together', 'moonshot', 'qianfan', 'nvidia', 'nanogpt', 'huggingface', 'venice', 'xiaomi', 'vllm', 'litellm', 'cloudflare-ai-gateway', 'qwen-portal', ...
1082
+ type OAuthProvider,
1085
1083
  type OAuthCredentials,
1086
1084
  } from "@f5-sales-demo/pi-ai";
1087
1085
  ```
1088
1086
 
1089
- `loginOpenAICodex` accepts an optional `originator` value used in the OAuth flow:
1090
-
1091
- ```typescript
1092
- await loginOpenAICodex({
1093
- onAuth: ({ url }) => console.log(url),
1094
- originator: "my-cli",
1095
- });
1096
- ```
1097
-
1098
1087
  ### Login Flow Example
1099
1088
 
1100
1089
  ```typescript
@@ -1149,7 +1138,7 @@ const response = await complete(
1149
1138
 
1150
1139
  ### Provider Notes
1151
1140
 
1152
- **OpenAI Codex**: Requires a ChatGPT Plus or Pro subscription. Provides access to GPT-5.x Codex models with extended context windows and reasoning capabilities. The library automatically handles session-based prompt caching when `sessionId` is provided in stream options.
1141
+ **OpenAI**: Set `OPENAI_API_KEY` for usage-based Platform API access. ChatGPT subscription access is available through the official Codex CLI, not this library.
1153
1142
 
1154
1143
  **GitHub Copilot**: If you get "The requested model is not supported" error, enable the model manually in VS Code: open Copilot Chat, click the model selector, select the model (warning icon), and click "Enable".
1155
1144
 
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "type": "module",
3
3
  "name": "@f5-sales-demo/pi-ai",
4
- "version": "20.19.1",
4
+ "version": "20.19.2",
5
5
  "description": "Unified LLM API with automatic model discovery and provider configuration",
6
6
  "homepage": "https://github.com/f5-sales-demo/xcsh",
7
7
  "author": "Can Boluk",
@@ -46,7 +46,7 @@
46
46
  "@anthropic-ai/sdk": "^0.115",
47
47
  "@aws-sdk/client-bedrock-runtime": "^3",
48
48
  "@bufbuild/protobuf": "^2.11",
49
- "@f5-sales-demo/pi-utils": "20.19.1",
49
+ "@f5-sales-demo/pi-utils": "20.19.2",
50
50
  "@google/genai": "^2.15",
51
51
  "@sinclair/typebox": "^0.34",
52
52
  "@smithy/node-http-handler": "^4.4",