diff --git a/README.md b/README.md index 2475e28b9..42d87c398 100644 --- a/README.md +++ b/README.md @@ -62,6 +62,7 @@ To chat with other community members you can join the [Agentuity Discord server] The structure of this mono repository: +- `packages/aigateway` - AI Gateway service client for model discovery and LLM completions - `packages/auth` - Agentuity unified Authentication package - `packages/claude-code` - Claude Code plugin with multi-agent coding team - `packages/cli` - the Agentuity command line tool diff --git a/apps/docs/scripts/generate-api-reference.ts b/apps/docs/scripts/generate-api-reference.ts index 74d7c91cb..813fc13b7 100644 --- a/apps/docs/scripts/generate-api-reference.ts +++ b/apps/docs/scripts/generate-api-reference.ts @@ -9,6 +9,7 @@ import type { Service, } from '../../../packages/core/src/services/api-reference.ts'; import { resolveFields } from '../../../packages/core/src/services/api-reference.ts'; +import aiGatewayService from '../../../packages/core/src/services/aigateway/api-reference.ts'; import apiKeysService from '../../../packages/core/src/services/apikey/api-reference.ts'; import coderService from '../../../packages/core/src/services/coder/api-reference.ts'; import databaseService from '../../../packages/core/src/services/db/api-reference.ts'; @@ -32,6 +33,7 @@ import vectorService from '../../../packages/core/src/services/vector/api-refere import webhooksService from '../../../packages/core/src/services/webhook/api-reference.ts'; const services: Service[] = [ + aiGatewayService, apiKeysService, coderService, databaseService, @@ -105,6 +107,16 @@ function renderResponseHeaders(headers: ResponseHeader[], subHeading: string): s ].join('\n'); } +function renderAuthentication(service: Service): string { + if (service.hasPublicEndpoints) { + if (service.slug !== 'ai-gateway') { + return 'Most requests require a Bearer token. Pass your API or SDK key in the `Authorization` header. Public endpoints (such as listing and fetching public snapshots) are noted below and do not require authentication.'; + } + return 'Most requests require a Bearer token. Pass your API or SDK key in the `Authorization` header. Public endpoints are noted below and do not require authentication.'; + } + return 'All requests require a Bearer token. Pass your API or SDK key in the `Authorization` header.'; +} + function renderEndpointSection(endpoint: Endpoint, headingLevel = 2, host?: string): string { const subHeading = '#'.repeat(headingLevel + 1); const pathParams = toParamTableInput(endpoint.pathParams, 'path'); @@ -180,6 +192,7 @@ function renderEndpointSection(endpoint: Endpoint, headingLevel = 2, host?: stri '', ``, '', + ...(endpoint.public ? ['**Authentication:** Public. No auth required.', ''] : []), paramSection, requestBodyParts.join('\n'), responseParts.join('\n'), @@ -222,7 +235,7 @@ description: ${service.description} ## Authentication -${service.hasPublicEndpoints ? 'Most requests require a Bearer token. Pass your API or SDK key in the `Authorization` header. Public endpoints (such as listing and fetching public snapshots) are noted below and do not require authentication.' : 'All requests require a Bearer token. Pass your API or SDK key in the `Authorization` header.'} +${renderAuthentication(service)} | Header | Value | |--------|-------| @@ -243,11 +256,17 @@ description: Direct HTTP access to Agentuity platform services {/* This file is auto-generated from Zod schemas. Do not edit manually. Run scripts/generate-api-reference.ts to regenerate. */} -import { Activity, Box, BrainCircuit, Building, Clock, Database, FolderKanban, Globe, HardDrive, Key, Layers, ListTodo, Mail, MessageSquare, Search, Server, Shield, Table, Timer, User, Webhook } from 'lucide-react'; +import { Activity, Box, BrainCircuit, Building, Clock, Cpu, Database, FolderKanban, Globe, HardDrive, Key, Layers, ListTodo, Mail, MessageSquare, Search, Server, Shield, Table, Timer, User, Webhook } from 'lucide-react'; Access any Agentuity Platform Service directly via REST APIs, the TypeScript SDK or the CLI. + } + /> + +## Authentication + +Most requests require a Bearer token. Pass your API or SDK key in the `Authorization` header. Public endpoints are noted below and do not require authentication. + +| Header | Value | +|--------|-------| +| `Authorization` | `Bearer YOUR_SDK_KEY` | + +You can find your SDK key in the [Agentuity Console](https://app.agentuity.com) under your project settings. + +--- + +## List Models + +List model metadata for LLM providers available through AI Gateway, grouped by provider. + + + +**Authentication:** Public. No auth required. + + + +### Response + +JSON response with provider keys mapped to arrays of supported model metadata under the `data` envelope. + +| Status | Description | +|--------|-------------| +| 200 | Model catalog returned. Public — no auth required. | + +### Response Fields + + + +### Example + + + +--- + +## Create Completion + +Create a completion through the AI Gateway auto-router. The gateway routes by model and request shape, so chat `messages` and legacy `prompt` payloads are both supported. + + + + + +### Request Body + +Completion request. Use `messages` for chat-compatible models and `prompt` for legacy OpenAI completions-compatible models. Additional provider-specific fields are passed through. + + + +### Response + +Provider-compatible completion response. + +| Status | Description | +|--------|-------------| +| 200 | Completion created | +| 400 | Invalid completion request | +| 401 | Unauthorized — invalid or missing API key | +| 402 | Payment required — upgrade to a paid plan | + +### Response Headers + +| Header | Description | +|--------|-------------| +| `X-Gateway-Cost` | Estimated total gateway cost in USD, when billing metadata is available. | +| `X-Gateway-Prompt-Tokens` | Prompt token count used for gateway billing. | +| `X-Gateway-Completion-Tokens` | Completion token count used for gateway billing. | + +### Response Fields + + + +### Example + + + +--- + +## Stream Completion + +Create a streaming completion through the AI Gateway auto-router. Set `stream: true` to receive Server-Sent Events token deltas. + + + + + +### Request Body + +Completion request with `stream` set to `true`. + + + +### Response + +Server-Sent Events stream. Each `data:` frame contains a provider-compatible delta payload. The stream ends with `data: [DONE]`. + +| Status | Description | +|--------|-------------| +| 200 | Streaming completion started | +| 400 | Invalid completion request | +| 401 | Unauthorized — invalid or missing API key | +| 402 | Payment required — upgrade to a paid plan | + +### Response Headers + +| Header | Description | +|--------|-------------| +| `Trailer` | Declares billing trailers such as `X-Gateway-Cost`, `X-Gateway-Prompt-Tokens`, and `X-Gateway-Completion-Tokens` for streamed responses. | +| `X-Gateway-Cost` | Estimated total gateway cost in USD. For streaming responses this may be delivered as an HTTP trailer after the body completes. | +| `X-Gateway-Prompt-Tokens` | Prompt token count used for gateway billing. For streaming responses this may be delivered as an HTTP trailer. | +| `X-Gateway-Completion-Tokens` | Completion token count used for gateway billing. For streaming responses this may be delivered as an HTTP trailer. | + +### Response Fields + + + +### Example + + + +--- diff --git a/apps/docs/src/web/content/reference/api/index.mdx b/apps/docs/src/web/content/reference/api/index.mdx index b762f319f..883bc9b1b 100644 --- a/apps/docs/src/web/content/reference/api/index.mdx +++ b/apps/docs/src/web/content/reference/api/index.mdx @@ -5,11 +5,17 @@ description: Direct HTTP access to Agentuity platform services {/* This file is auto-generated from Zod schemas. Do not edit manually. Run scripts/generate-api-reference.ts to regenerate. */} -import { Activity, Box, BrainCircuit, Building, Clock, Database, FolderKanban, Globe, HardDrive, Key, Layers, ListTodo, Mail, MessageSquare, Search, Server, Shield, Table, Timer, User, Webhook } from 'lucide-react'; +import { Activity, Box, BrainCircuit, Building, Clock, Cpu, Database, FolderKanban, Globe, HardDrive, Key, Layers, ListTodo, Mail, MessageSquare, Search, Server, Shield, Table, Timer, User, Webhook } from 'lucide-react'; Access any Agentuity Platform Service directly via REST APIs, the TypeScript SDK or the CLI. + } + /> DocsReferenceApiRouteRoute, } as any); +const DocsReferenceApiAiGatewayRoute = DocsReferenceApiAiGatewayRouteImport.update({ + id: '/ai-gateway', + path: '/ai-gateway', + getParentRoute: () => DocsReferenceApiRouteRoute, +} as any); const DocsCookbookTutorialsUnderstandingAgentsRoute = DocsCookbookTutorialsUnderstandingAgentsRouteImport.update({ id: '/cookbook/tutorials/understanding-agents', @@ -1281,6 +1287,7 @@ export interface FileRoutesByFullPath { '/cookbook/patterns/webhook-handler': typeof DocsCookbookPatternsWebhookHandlerRoute; '/cookbook/tutorials/rag-agent': typeof DocsCookbookTutorialsRagAgentRoute; '/cookbook/tutorials/understanding-agents': typeof DocsCookbookTutorialsUnderstandingAgentsRoute; + '/reference/api/ai-gateway': typeof DocsReferenceApiAiGatewayRoute; '/reference/api/api-keys': typeof DocsReferenceApiApiKeysRoute; '/reference/api/coder': typeof DocsReferenceApiCoderRoute; '/reference/api/database': typeof DocsReferenceApiDatabaseRoute; @@ -1468,6 +1475,7 @@ export interface FileRoutesByTo { '/cookbook/patterns/webhook-handler': typeof DocsCookbookPatternsWebhookHandlerRoute; '/cookbook/tutorials/rag-agent': typeof DocsCookbookTutorialsRagAgentRoute; '/cookbook/tutorials/understanding-agents': typeof DocsCookbookTutorialsUnderstandingAgentsRoute; + '/reference/api/ai-gateway': typeof DocsReferenceApiAiGatewayRoute; '/reference/api/api-keys': typeof DocsReferenceApiApiKeysRoute; '/reference/api/coder': typeof DocsReferenceApiCoderRoute; '/reference/api/database': typeof DocsReferenceApiDatabaseRoute; @@ -1659,6 +1667,7 @@ export interface FileRoutesById { '/_docs/cookbook/patterns/webhook-handler': typeof DocsCookbookPatternsWebhookHandlerRoute; '/_docs/cookbook/tutorials/rag-agent': typeof DocsCookbookTutorialsRagAgentRoute; '/_docs/cookbook/tutorials/understanding-agents': typeof DocsCookbookTutorialsUnderstandingAgentsRoute; + '/_docs/reference/api/ai-gateway': typeof DocsReferenceApiAiGatewayRoute; '/_docs/reference/api/api-keys': typeof DocsReferenceApiApiKeysRoute; '/_docs/reference/api/coder': typeof DocsReferenceApiCoderRoute; '/_docs/reference/api/database': typeof DocsReferenceApiDatabaseRoute; @@ -1850,6 +1859,7 @@ export interface FileRouteTypes { | '/cookbook/patterns/webhook-handler' | '/cookbook/tutorials/rag-agent' | '/cookbook/tutorials/understanding-agents' + | '/reference/api/ai-gateway' | '/reference/api/api-keys' | '/reference/api/coder' | '/reference/api/database' @@ -2037,6 +2047,7 @@ export interface FileRouteTypes { | '/cookbook/patterns/webhook-handler' | '/cookbook/tutorials/rag-agent' | '/cookbook/tutorials/understanding-agents' + | '/reference/api/ai-gateway' | '/reference/api/api-keys' | '/reference/api/coder' | '/reference/api/database' @@ -2227,6 +2238,7 @@ export interface FileRouteTypes { | '/_docs/cookbook/patterns/webhook-handler' | '/_docs/cookbook/tutorials/rag-agent' | '/_docs/cookbook/tutorials/understanding-agents' + | '/_docs/reference/api/ai-gateway' | '/_docs/reference/api/api-keys' | '/_docs/reference/api/coder' | '/_docs/reference/api/database' @@ -3455,6 +3467,13 @@ declare module '@tanstack/react-router' { preLoaderRoute: typeof DocsReferenceApiApiKeysRouteImport; parentRoute: typeof DocsReferenceApiRouteRoute; }; + '/_docs/reference/api/ai-gateway': { + id: '/_docs/reference/api/ai-gateway'; + path: '/ai-gateway'; + fullPath: '/reference/api/ai-gateway'; + preLoaderRoute: typeof DocsReferenceApiAiGatewayRouteImport; + parentRoute: typeof DocsReferenceApiRouteRoute; + }; '/_docs/cookbook/tutorials/understanding-agents': { id: '/_docs/cookbook/tutorials/understanding-agents'; path: '/cookbook/tutorials/understanding-agents'; @@ -3655,6 +3674,7 @@ declare module '@tanstack/react-router' { } interface DocsReferenceApiRouteRouteChildren { + DocsReferenceApiAiGatewayRoute: typeof DocsReferenceApiAiGatewayRoute; DocsReferenceApiApiKeysRoute: typeof DocsReferenceApiApiKeysRoute; DocsReferenceApiCoderRoute: typeof DocsReferenceApiCoderRoute; DocsReferenceApiDatabaseRoute: typeof DocsReferenceApiDatabaseRoute; @@ -3681,6 +3701,7 @@ interface DocsReferenceApiRouteRouteChildren { } const DocsReferenceApiRouteRouteChildren: DocsReferenceApiRouteRouteChildren = { + DocsReferenceApiAiGatewayRoute: DocsReferenceApiAiGatewayRoute, DocsReferenceApiApiKeysRoute: DocsReferenceApiApiKeysRoute, DocsReferenceApiCoderRoute: DocsReferenceApiCoderRoute, DocsReferenceApiDatabaseRoute: DocsReferenceApiDatabaseRoute, diff --git a/apps/docs/src/web/routes/_docs/reference/api/ai-gateway.tsx b/apps/docs/src/web/routes/_docs/reference/api/ai-gateway.tsx new file mode 100644 index 000000000..ab125c685 --- /dev/null +++ b/apps/docs/src/web/routes/_docs/reference/api/ai-gateway.tsx @@ -0,0 +1,7 @@ +import { createFileRoute } from '@tanstack/react-router'; +import { MDXPage } from '../../../../components/docs/mdx-page'; + +export const Route = createFileRoute('/_docs/reference/api/ai-gateway')({ + component: () => , + staticData: { crumb: 'AI Gateway' }, +}); diff --git a/bun.lock b/bun.lock index 724e443e9..7c7757a0a 100644 --- a/bun.lock +++ b/bun.lock @@ -418,6 +418,21 @@ "vite": "^7.2.7", }, }, + "packages/aigateway": { + "name": "@agentuity/aigateway", + "version": "2.0.14", + "dependencies": { + "@agentuity/core": "workspace:*", + "@agentuity/server": "workspace:*", + "zod": "^4.3.5", + }, + "devDependencies": { + "@types/bun": "latest", + "@types/node": "^22.0.0", + "bun-types": "latest", + "typescript": "^5.9.0", + }, + }, "packages/auth": { "name": "@agentuity/auth", "version": "2.0.14", @@ -520,6 +535,8 @@ "name": "@agentuity/coder-tui", "version": "2.0.14", "dependencies": { + "@agentuity/core": "workspace:*", + "@agentuity/server": "workspace:*", "@mariozechner/pi-coding-agent": "^0.72.1", "@mariozechner/pi-tui": "^0.72.1", "@sinclair/typebox": "^0.34.49", @@ -967,6 +984,8 @@ "@adobe/css-tools": ["@adobe/css-tools@4.4.4", "", {}, "sha512-Elp+iwUx5rN5+Y8xLt5/GRoG20WGoDCQ/1Fb+1LiGtvwbDavuSk0jhD/eZdckHAuzcDzccnkv+rEjyWfRx18gg=="], + "@agentuity/aigateway": ["@agentuity/aigateway@workspace:packages/aigateway"], + "@agentuity/auth": ["@agentuity/auth@workspace:packages/auth"], "@agentuity/claude-code": ["@agentuity/claude-code@workspace:packages/claude-code"], @@ -3001,7 +3020,7 @@ "function-bind": ["function-bind@1.1.2", "", {}, "sha512-7XHNxH7qX9xG5mIwxkhumTox/MIRNcOgDrxWsMt2pAr23WHp6MrRlN7FBSFpCpr+oVO0F744iUgR82nJMfG2SA=="], - "gaxios": ["gaxios@7.1.4", "", { "dependencies": { "extend": "^3.0.2", "https-proxy-agent": "^7.0.1", "node-fetch": "^3.3.2" } }, "sha512-bTIgTsM2bWn3XklZISBTQX7ZSddGW+IO3bMdGaemHZ3tbqExMENHLx6kKZ/KlejgrMtj8q7wBItt51yegqalrA=="], + "gaxios": ["gaxios@6.7.1", "", { "dependencies": { "extend": "^3.0.2", "https-proxy-agent": "^7.0.1", "is-stream": "^2.0.0", "node-fetch": "^2.6.9", "uuid": "^9.0.1" } }, "sha512-LDODD4TMYx7XXdpwxAVRAIAuB0bzv0s+ywFonY46k126qzQHT9ygyoa9tncmOiQmmDrik65UYsEkv3lbfqQ3yQ=="], "gcp-metadata": ["gcp-metadata@6.1.1", "", { "dependencies": { "gaxios": "^6.1.1", "google-logging-utils": "^0.0.2", "json-bigint": "^1.0.0" } }, "sha512-a4tiq7E0/5fTjxPAaH4jpjkSv/uCaU2p5KC6HVGrvl0cDjA8iBZv4vv1gyzlmK0ZUKqwpOyQMKzZQe3lTit77A=="], @@ -3047,7 +3066,7 @@ "google-auth-library": ["google-auth-library@10.6.2", "", { "dependencies": { "base64-js": "^1.3.0", "ecdsa-sig-formatter": "^1.0.11", "gaxios": "^7.1.4", "gcp-metadata": "8.1.2", "google-logging-utils": "1.1.3", "jws": "^4.0.0" } }, "sha512-e27Z6EThmVNNvtYASwQxose/G57rkRuaRbQyxM2bvYLLX/GqWZ5chWq2EBoUchJbCc57eC9ArzO5wMsEmWftCw=="], - "google-logging-utils": ["google-logging-utils@1.1.3", "", {}, "sha512-eAmLkjDjAFCVXg7A1unxHsLf961m6y17QFqXqAXGj/gVkKFrEICfStRfwUlGNfeCEjNRa32JEWOUTlYXPyyKvA=="], + "google-logging-utils": ["google-logging-utils@0.0.2", "", {}, "sha512-NEgUnEcBiP5HrPzufUkBzJOD/Sxsco3rLNo1F1TNf7ieU8ryUzBhqba8r756CjLX7rn3fHl6iLEwPYuqpoKgQQ=="], "gopd": ["gopd@1.2.0", "", {}, "sha512-ZUKRh6/kUFoAiTAtTYPZJ3hw9wNxx+BIBOijnlG9PnrJsCcSjs1wyyD6vJpaYtgnzDrKYRSqf3OO6Rfa93xsRg=="], @@ -3129,7 +3148,7 @@ "husky": ["husky@9.1.7", "", { "bin": { "husky": "bin.js" } }, "sha512-5gs5ytaNjBrh5Ow3zrvdUUY+0VxIuWVL4i9irt6friV+BqdCfmV11CQTWMiBYWHbXhco+J1kHfTOUkePhCDvMA=="], - "iconv-lite": ["iconv-lite@0.6.3", "", { "dependencies": { "safer-buffer": ">= 2.1.2 < 3.0.0" } }, "sha512-4fCk79wshMdzMp2rH06qWrJE4iolqLhCUH+OiuIgU++RB0+94NlDL81atO7GX55uUKueo0txHNtvEyI6D7WdMw=="], + "iconv-lite": ["iconv-lite@0.7.2", "", { "dependencies": { "safer-buffer": ">= 2.1.2 < 3.0.0" } }, "sha512-im9DjEDQ55s9fL4EYzOAv0yMqmMBSZp6G0VvFyTMPKWxiSBHUj9NW/qqLmXUwXrrM7AvqSlTCfvqRb0cM8yYqw=="], "ieee754": ["ieee754@1.2.1", "", {}, "sha512-dcyqhDvX1C46lXZcVqCpK+FtMRQVdIMN6/Df5js2zouUsqG7I6sFxitIC+7KYK29KdXOLHdu9zL4sFnoVQnqaA=="], @@ -3553,7 +3572,7 @@ "node-domexception": ["node-domexception@1.0.0", "", {}, "sha512-/jKZoMpw0F8GRwl4/eLROPA3cfcXtLApP0QzLmUT/HuPCZWyB7IY9ZrMeKw2O/nFIqPQB3PVM9aYm0F312AXDQ=="], - "node-fetch": ["node-fetch@3.3.2", "", { "dependencies": { "data-uri-to-buffer": "^4.0.0", "fetch-blob": "^3.1.4", "formdata-polyfill": "^4.0.10" } }, "sha512-dRB78srN/l6gqWulah9SrxeYnxeddIG30+GOqK/9OlLVyLg3HPnr6SqOWTWOXKRwC2eGYCkZ59NNuSgvSrpgOA=="], + "node-fetch": ["node-fetch@2.7.0", "", { "dependencies": { "whatwg-url": "^5.0.0" }, "peerDependencies": { "encoding": "^0.1.0" }, "optionalPeers": ["encoding"] }, "sha512-c4FRfUm/dbcWZ7U+1Wq0AwCyFL+3nt2bEw05wfxSz+DWpWsitgmSgYmy2dQdWyKC1694ELPqMs/YzUSNozLt8A=="], "node-gyp-build-optional-packages": ["node-gyp-build-optional-packages@5.2.2", "", { "dependencies": { "detect-libc": "^2.0.1" }, "bin": { "node-gyp-build-optional-packages": "bin.js", "node-gyp-build-optional-packages-optional": "optional.js", "node-gyp-build-optional-packages-test": "build-test.js" } }, "sha512-s+w+rBWnpTMwSFbaE0UXsRlg7hU4FjekKU4eyAih5T8nJuNZT1nNsskXpxmeqSK9UzkBl6UgRlnKc8hz8IEqOw=="], @@ -3669,7 +3688,7 @@ "picomatch": ["picomatch@4.0.4", "", {}, "sha512-QP88BAKvMam/3NxH6vj2o21R6MjxZUAd6nlwAS/pnGvN9IVLocLHxGYIzFhg6fUQ+5th6P4dv4eW9jX3DSIj7A=="], - "pkg-types": ["pkg-types@1.3.1", "", { "dependencies": { "confbox": "^0.1.8", "mlly": "^1.7.4", "pathe": "^2.0.1" } }, "sha512-/Jm5M4RvtBFVkKWRu2BLUTNP8/M2a+UwuAX+ae4770q1qVGtfjG+WTCupoZixokjmHiry8uI+dlY8KXYV5HVVQ=="], + "pkg-types": ["pkg-types@2.3.1", "", { "dependencies": { "confbox": "^0.2.4", "exsolve": "^1.0.8", "pathe": "^2.0.3" } }, "sha512-y+ichcgc2LrADuhLNAx8DFjVfgz91pRxfZdI3UDhxHvcVEZsenLO+7XaU5vOp0u/7V/wZ+plyuQxtrDlZJ+yeg=="], "playwright": ["playwright@1.59.1", "", { "dependencies": { "playwright-core": "1.59.1" }, "optionalDependencies": { "fsevents": "2.3.2" }, "bin": { "playwright": "cli.js" } }, "sha512-C8oWjPR3F81yljW9o5OxcWzfh6avkVwDD2VYdwIGqTkl+OGFISgypqzfu7dOe4QNLL2aqcWBmI3PMtLIK233lw=="], @@ -4523,8 +4542,6 @@ "c12/chokidar": ["chokidar@5.0.0", "", { "dependencies": { "readdirp": "^5.0.0" } }, "sha512-TQMmc3w+5AxjpL8iIiwebF73dRDF4fBIieAqGn9RGCWaEVwQ6Fb2cGe31Yns0RRIzii5goJ1Y7xbMwo1TxMplw=="], - "c12/pkg-types": ["pkg-types@2.3.1", "", { "dependencies": { "confbox": "^0.2.4", "exsolve": "^1.0.8", "pathe": "^2.0.3" } }, "sha512-y+ichcgc2LrADuhLNAx8DFjVfgz91pRxfZdI3UDhxHvcVEZsenLO+7XaU5vOp0u/7V/wZ+plyuQxtrDlZJ+yeg=="], - "chalk/ansi-styles": ["ansi-styles@4.3.0", "", { "dependencies": { "color-convert": "^2.0.1" } }, "sha512-zbB9rCJAT1rbjiVDb2hqKFHNYLxgtk8NURxZ3IZwD3F6NtxbXZQCnnSi1Lkx+IDohdPlFp222wVALIheZJQSEg=="], "chalk/supports-color": ["supports-color@7.2.0", "", { "dependencies": { "has-flag": "^4.0.0" } }, "sha512-qpCAvRl9stuOHveKsn7HncJRvv501qIacKzQlO/+Lwxc9+0q2wLyv4Dfvt80/DPn2pqOBsJdDiogXGR9+OvwRw=="], @@ -4547,6 +4564,8 @@ "cytoscape-fcose/cose-base": ["cose-base@2.2.0", "", { "dependencies": { "layout-base": "^2.0.0" } }, "sha512-AzlgcsCbUMymkADOJtQm3wO9S3ltPfYOFD5033keQn9NJzIbtnZj+UdBJe7DYml/8TdbtHJW3j58SOnKhWY/5g=="], + "d3-dsv/iconv-lite": ["iconv-lite@0.6.3", "", { "dependencies": { "safer-buffer": ">= 2.1.2 < 3.0.0" } }, "sha512-4fCk79wshMdzMp2rH06qWrJE4iolqLhCUH+OiuIgU++RB0+94NlDL81atO7GX55uUKueo0txHNtvEyI6D7WdMw=="], + "d3-sankey/d3-array": ["d3-array@2.12.1", "", { "dependencies": { "internmap": "^1.0.0" } }, "sha512-B0ErZK/66mHtEsR1TkPEEkwdy+WDesimkM5gpZr5Dsg54BiTA5RXtYW5qTLIAcekaS9xfZrzBLF/OAkB3Qn1YQ=="], "d3-sankey/d3-shape": ["d3-shape@1.3.7", "", { "dependencies": { "d3-path": "1" } }, "sha512-EUkvKjqPFUAZyOlhY5gzCxCeI0Aep04LwIRpsZ/mLFelJiUfnK56jo5JMDSE7yyP2kLSb6LtF+S5chMk7uqPqw=="], @@ -4571,6 +4590,8 @@ "effect/uuid": ["uuid@13.0.2", "", { "bin": { "uuid": "dist-node/bin/uuid" } }, "sha512-vzi9uRZ926x4XV73S/4qQaTwPXM2JBj6/6lI/byHH1jOpCzb0zDbfytgA9LcN/hzb2l7WQSQnxITOVx5un/wGw=="], + "encoding-sniffer/iconv-lite": ["iconv-lite@0.6.3", "", { "dependencies": { "safer-buffer": ">= 2.1.2 < 3.0.0" } }, "sha512-4fCk79wshMdzMp2rH06qWrJE4iolqLhCUH+OiuIgU++RB0+94NlDL81atO7GX55uUKueo0txHNtvEyI6D7WdMw=="], + "escodegen/source-map": ["source-map@0.6.1", "", {}, "sha512-UjgapumWlbMhkBgzT7Ykc5YXUT46F0iKu8SGXq0bcwP5dz/h0Plj6enJqjz1Zbq2l5WaqYnrVbwWOWMyF3F47g=="], "extract-zip/yauzl": ["yauzl@2.10.0", "", { "dependencies": { "buffer-crc32": "~0.2.3", "fd-slicer": "~1.1.0" } }, "sha512-p4a9I6X6nu6IhoGmBqAcbJy1mlC4j27vEPZX9F4L4/vZT3Lyq1VkFHw/V/PUcB9Buo+DG3iHkT0x3Qya58zc3g=="], @@ -4579,12 +4600,14 @@ "form-data/mime-types": ["mime-types@2.1.35", "", { "dependencies": { "mime-db": "1.52.0" } }, "sha512-ZDY+bPm5zTTF+YpCrAU9nK0UgICYPT0QtT1NZWFv4s++TNkcgVaT0g6+4R2uI4MjQjzysHB1zxuWL50hzaeXiw=="], - "gcp-metadata/gaxios": ["gaxios@6.7.1", "", { "dependencies": { "extend": "^3.0.2", "https-proxy-agent": "^7.0.1", "is-stream": "^2.0.0", "node-fetch": "^2.6.9", "uuid": "^9.0.1" } }, "sha512-LDODD4TMYx7XXdpwxAVRAIAuB0bzv0s+ywFonY46k126qzQHT9ygyoa9tncmOiQmmDrik65UYsEkv3lbfqQ3yQ=="], + "gaxios/uuid": ["uuid@9.0.1", "", { "bin": { "uuid": "dist/bin/uuid" } }, "sha512-b+1eJOlsR9K8HJpow9Ok3fiWOWSIcIzXodvv0rQjVoOVNpWMpxf1wZNpt4y9h10odCNrqnYp1OBzRktckBe3sA=="], - "gcp-metadata/google-logging-utils": ["google-logging-utils@0.0.2", "", {}, "sha512-NEgUnEcBiP5HrPzufUkBzJOD/Sxsco3rLNo1F1TNf7ieU8ryUzBhqba8r756CjLX7rn3fHl6iLEwPYuqpoKgQQ=="], + "google-auth-library/gaxios": ["gaxios@7.1.4", "", { "dependencies": { "extend": "^3.0.2", "https-proxy-agent": "^7.0.1", "node-fetch": "^3.3.2" } }, "sha512-bTIgTsM2bWn3XklZISBTQX7ZSddGW+IO3bMdGaemHZ3tbqExMENHLx6kKZ/KlejgrMtj8q7wBItt51yegqalrA=="], "google-auth-library/gcp-metadata": ["gcp-metadata@8.1.2", "", { "dependencies": { "gaxios": "^7.0.0", "google-logging-utils": "^1.0.0", "json-bigint": "^1.0.0" } }, "sha512-zV/5HKTfCeKWnxG0Dmrw51hEWFGfcF2xiXqcA3+J90WDuP0SvoiSO5ORvcBsifmx/FoIjgQN3oNOGaQ5PhLFkg=="], + "google-auth-library/google-logging-utils": ["google-logging-utils@1.1.3", "", {}, "sha512-eAmLkjDjAFCVXg7A1unxHsLf961m6y17QFqXqAXGj/gVkKFrEICfStRfwUlGNfeCEjNRa32JEWOUTlYXPyyKvA=="], + "happy-dom/whatwg-mimetype": ["whatwg-mimetype@3.0.0", "", {}, "sha512-nt+N2dzIutVRxARx1nghPKGv1xHikU7HKdfafKkLNLindmPU/ch3U31NOCGGA/dmPcmb1VlofO0vnKAcsm0o/Q=="], "hast-util-from-html/parse5": ["parse5@7.3.0", "", { "dependencies": { "entities": "^6.0.0" } }, "sha512-IInvU7fabl34qmi9gY8XOVxhYyMyuH2xUNpb2q8/Y+7552KlejkRvqvD19nMoUW/uQGGbqNpA6Tufu5FL5BZgw=="], @@ -4617,15 +4640,15 @@ "micromatch/picomatch": ["picomatch@2.3.2", "", {}, "sha512-V7+vQEJ06Z+c5tSye8S+nHUfI51xoXIXjHQ99cQtKUkQqqO1kO/KCJUfZXuB47h/YBlDhah2H3hdUGXn8ie0oA=="], - "mongodb-connection-string-url/whatwg-url": ["whatwg-url@14.2.0", "", { "dependencies": { "tr46": "^5.1.0", "webidl-conversions": "^7.0.0" } }, "sha512-De72GdQZzNTUBBChsXueQUnPKDkg/5A5zp7pFDuQAj5UFoENpiACU0wlCvzpAGnTkj++ihpKwKyYewn/XNUbKw=="], + "mlly/pkg-types": ["pkg-types@1.3.1", "", { "dependencies": { "confbox": "^0.1.8", "mlly": "^1.7.4", "pathe": "^2.0.1" } }, "sha512-/Jm5M4RvtBFVkKWRu2BLUTNP8/M2a+UwuAX+ae4770q1qVGtfjG+WTCupoZixokjmHiry8uI+dlY8KXYV5HVVQ=="], - "mysql2/iconv-lite": ["iconv-lite@0.7.2", "", { "dependencies": { "safer-buffer": ">= 2.1.2 < 3.0.0" } }, "sha512-im9DjEDQ55s9fL4EYzOAv0yMqmMBSZp6G0VvFyTMPKWxiSBHUj9NW/qqLmXUwXrrM7AvqSlTCfvqRb0cM8yYqw=="], + "mongodb-connection-string-url/whatwg-url": ["whatwg-url@14.2.0", "", { "dependencies": { "tr46": "^5.1.0", "webidl-conversions": "^7.0.0" } }, "sha512-De72GdQZzNTUBBChsXueQUnPKDkg/5A5zp7pFDuQAj5UFoENpiACU0wlCvzpAGnTkj++ihpKwKyYewn/XNUbKw=="], "next/postcss": ["postcss@8.4.31", "", { "dependencies": { "nanoid": "^3.3.6", "picocolors": "^1.0.0", "source-map-js": "^1.0.2" } }, "sha512-PS08Iboia9mts/2ygV3eLpY5ghnUcfLV/EXTOW1E2qYxJKGGBUtNjN76FYHnMs36RmARn41bC0AZmn+rR0OVpQ=="], "nextjs-app-agentuity/@vitejs/plugin-react": ["@vitejs/plugin-react@4.7.0", "", { "dependencies": { "@babel/core": "^7.28.0", "@babel/plugin-transform-react-jsx-self": "^7.27.1", "@babel/plugin-transform-react-jsx-source": "^7.27.1", "@rolldown/pluginutils": "1.0.0-beta.27", "@types/babel__core": "^7.20.5", "react-refresh": "^0.17.0" }, "peerDependencies": { "vite": "^4.2.0 || ^5.0.0 || ^6.0.0 || ^7.0.0" } }, "sha512-gUu9hwfWvvEDBBmgtAowQCojwZmJ5mcLn3aufeCsitijs3+f2NsrPtlAWIR6OPiqljl96GVCUbLe0HyqIpVaoA=="], - "node-fetch/data-uri-to-buffer": ["data-uri-to-buffer@4.0.1", "", {}, "sha512-0R9ikRb668HB7QDxT1vkpuUBtqc53YyAwMwGeUFKRojY/NWKvdZ+9UYtRfGmhqNbRkTSVpMbmyhXipFFv2cb/A=="], + "node-fetch/whatwg-url": ["whatwg-url@5.0.0", "", { "dependencies": { "tr46": "~0.0.3", "webidl-conversions": "^3.0.0" } }, "sha512-saE57nupxk6v3HY35+jzBwYa0rKSy0XR8JSxZPwgLr7ys0IBzhGviA1/TUGJLmSVqs8pb9AnvICXEuOHLprYTw=="], "normalize-package-data/hosted-git-info": ["hosted-git-info@7.0.2", "", { "dependencies": { "lru-cache": "^10.0.1" } }, "sha512-puUZAUKT5m8Zzvs72XWy3HtvVbTWljRE66cP60bxJzAqf2DgICo7lYTY2IHUmLnNpjYvw5bvmoHvPc0QO2a62w=="], @@ -4645,8 +4668,6 @@ "path-scurry/lru-cache": ["lru-cache@11.3.6", "", {}, "sha512-Gf/KoL3C/MlI7Bt0PGI9I+TeTC/I6r/csU58N4BSNc4lppLBeKsOdFYkK+dX0ABDUMJNfCHTyPpzwwO21Awd3A=="], - "pkg-types/confbox": ["confbox@0.1.8", "", {}, "sha512-RMtmw0iFkeR4YV+fUOSucriAQNb9g8zFR52MWCtl+cCZOFRNL6zeB395vPzFhEjjn4fMxXudmELnl/KF/WrK6w=="], - "postcss/nanoid": ["nanoid@3.3.12", "", { "bin": { "nanoid": "bin/nanoid.cjs" } }, "sha512-ZB9RH/39qpq5Vu6Y+NmUaFhQR6pp+M2Xt76XBnEwDaGcVAqhlvxrl3B2bKS5D3NH3QR76v3aSrKaF/Kiy7lEtQ=="], "prebuild-install/tar-fs": ["tar-fs@2.1.4", "", { "dependencies": { "chownr": "^1.1.1", "mkdirp-classic": "^0.5.2", "pump": "^3.0.0", "tar-stream": "^2.1.4" } }, "sha512-mDAjwmZdh7LTT6pNleZ05Yt65HC3E+NiQzl672vQG38jIrehtJk/J3mNwIg+vShQPcLF/LV7CMnDW6vjj6sfYQ=="], @@ -4709,6 +4730,8 @@ "webrtc-test/@vitejs/plugin-react": ["@vitejs/plugin-react@4.7.0", "", { "dependencies": { "@babel/core": "^7.28.0", "@babel/plugin-transform-react-jsx-self": "^7.27.1", "@babel/plugin-transform-react-jsx-source": "^7.27.1", "@rolldown/pluginutils": "1.0.0-beta.27", "@types/babel__core": "^7.20.5", "react-refresh": "^0.17.0" }, "peerDependencies": { "vite": "^4.2.0 || ^5.0.0 || ^6.0.0 || ^7.0.0" } }, "sha512-gUu9hwfWvvEDBBmgtAowQCojwZmJ5mcLn3aufeCsitijs3+f2NsrPtlAWIR6OPiqljl96GVCUbLe0HyqIpVaoA=="], + "whatwg-encoding/iconv-lite": ["iconv-lite@0.6.3", "", { "dependencies": { "safer-buffer": ">= 2.1.2 < 3.0.0" } }, "sha512-4fCk79wshMdzMp2rH06qWrJE4iolqLhCUH+OiuIgU++RB0+94NlDL81atO7GX55uUKueo0txHNtvEyI6D7WdMw=="], + "wrap-ansi/ansi-styles": ["ansi-styles@4.3.0", "", { "dependencies": { "color-convert": "^2.0.1" } }, "sha512-zbB9rCJAT1rbjiVDb2hqKFHNYLxgtk8NURxZ3IZwD3F6NtxbXZQCnnSi1Lkx+IDohdPlFp222wVALIheZJQSEg=="], "wrap-ansi-cjs/ansi-styles": ["ansi-styles@4.3.0", "", { "dependencies": { "color-convert": "^2.0.1" } }, "sha512-zbB9rCJAT1rbjiVDb2hqKFHNYLxgtk8NURxZ3IZwD3F6NtxbXZQCnnSi1Lkx+IDohdPlFp222wVALIheZJQSEg=="], @@ -4869,9 +4892,7 @@ "form-data/mime-types/mime-db": ["mime-db@1.52.0", "", {}, "sha512-sPU4uV7dYlvtWJxwwxHD0PuihVNiE7TyAbQ5SWxDCB9mUYvOgroQOwYQQOKPJ8CIbE+1ETVlOoK1UC2nU3gYvg=="], - "gcp-metadata/gaxios/node-fetch": ["node-fetch@2.7.0", "", { "dependencies": { "whatwg-url": "^5.0.0" }, "peerDependencies": { "encoding": "^0.1.0" }, "optionalPeers": ["encoding"] }, "sha512-c4FRfUm/dbcWZ7U+1Wq0AwCyFL+3nt2bEw05wfxSz+DWpWsitgmSgYmy2dQdWyKC1694ELPqMs/YzUSNozLt8A=="], - - "gcp-metadata/gaxios/uuid": ["uuid@9.0.1", "", { "bin": { "uuid": "dist/bin/uuid" } }, "sha512-b+1eJOlsR9K8HJpow9Ok3fiWOWSIcIzXodvv0rQjVoOVNpWMpxf1wZNpt4y9h10odCNrqnYp1OBzRktckBe3sA=="], + "google-auth-library/gaxios/node-fetch": ["node-fetch@3.3.2", "", { "dependencies": { "data-uri-to-buffer": "^4.0.0", "fetch-blob": "^3.1.4", "formdata-polyfill": "^4.0.10" } }, "sha512-dRB78srN/l6gqWulah9SrxeYnxeddIG30+GOqK/9OlLVyLg3HPnr6SqOWTWOXKRwC2eGYCkZ59NNuSgvSrpgOA=="], "hast-util-from-html/parse5/entities": ["entities@6.0.1", "", {}, "sha512-aN97NXWF6AWBTahfVOIrB/NShkzi5H7F9r1s9mD3cDj4Ko5f2qhhVoYMibXF7GlLveb/D2ioWay8lxI97Ven3g=="], @@ -4885,6 +4906,8 @@ "lazystream/readable-stream/string_decoder": ["string_decoder@1.1.1", "", { "dependencies": { "safe-buffer": "~5.1.0" } }, "sha512-n/ShnvDi6FHbbVfviro+WojiFzv+s8MPMHBczVePfUpDJLwoLT0ht1l4YwBCbi8pJAveEEdnkHyPyTP/mzRfwg=="], + "mlly/pkg-types/confbox": ["confbox@0.1.8", "", {}, "sha512-RMtmw0iFkeR4YV+fUOSucriAQNb9g8zFR52MWCtl+cCZOFRNL6zeB395vPzFhEjjn4fMxXudmELnl/KF/WrK6w=="], + "mongodb-connection-string-url/whatwg-url/tr46": ["tr46@5.1.1", "", { "dependencies": { "punycode": "^2.3.1" } }, "sha512-hdF5ZgjTqgAntKkklYw0R03MG2x/bSzTtkxmIRw/sTNV8YXsCJ1tfLAX23lhxhHJlEf3CRCOCGGWw3vI3GaSPw=="], "mongodb-connection-string-url/whatwg-url/webidl-conversions": ["webidl-conversions@7.0.0", "", {}, "sha512-VwddBukDzu71offAQR975unBIGqfKZpM+8ZX6ySk8nYhVoo5CYaZyzt3YBvYtRtO+aoGlqxPg/B87NGVZ/fu6g=="], @@ -4895,6 +4918,10 @@ "nextjs-app-agentuity/@vitejs/plugin-react/react-refresh": ["react-refresh@0.17.0", "", {}, "sha512-z6F7K9bV85EfseRCp2bzrpyQ0Gkw1uLoCel9XBVWPg/TjRj94SkJzUTGfOa4bs7iJvBWtQG0Wq7wnI0syw3EBQ=="], + "node-fetch/whatwg-url/tr46": ["tr46@0.0.3", "", {}, "sha512-N3WMsuqV66lT30CrXNbEjx4GEwlow3v6rr4mCcv6prnfwhS01rkgyFdjPNBYd9br7LpXV1+Emh01fHnq2Gdgrw=="], + + "node-fetch/whatwg-url/webidl-conversions": ["webidl-conversions@3.0.1", "", {}, "sha512-2JAn3z8AR6rjK8Sm8orRC0h/bcl/DqL7tRPdGZ4I1CjdF+EaMLmYxBHyXuKL849eucPFhvBoxMsflfOb8kxaeQ=="], + "normalize-package-data/hosted-git-info/lru-cache": ["lru-cache@10.4.3", "", {}, "sha512-JNAzZcXrCt42VGLuYz0zfAzDfAvJWW6AfYlDBQyDV5DClI2m5sAmK+OIO7s59XfsRsWHp02jAJrRadPRGTt6SQ=="], "oauth/@vitejs/plugin-react/@rolldown/pluginutils": ["@rolldown/pluginutils@1.0.0-beta.27", "", {}, "sha512-+d0F4MKMCbeVUJwG96uQ4SgAznZNSq93I3V+9NHA4OpvqG8mRCpGdKmK8l/dl02h2CCDHwW2FqilnTyDcAnqjA=="], @@ -5111,7 +5138,7 @@ "docs/ai/@ai-sdk/gateway/@vercel/oidc": ["@vercel/oidc@3.1.0", "", {}, "sha512-Fw28YZpRnA3cAHHDlkt7xQHiJ0fcL+NRcIqsocZQUSmbzeIKRpwttJjik5ZGanXP+vlA4SbTg+AbA3bP363l+w=="], - "gcp-metadata/gaxios/node-fetch/whatwg-url": ["whatwg-url@5.0.0", "", { "dependencies": { "tr46": "~0.0.3", "webidl-conversions": "^3.0.0" } }, "sha512-saE57nupxk6v3HY35+jzBwYa0rKSy0XR8JSxZPwgLr7ys0IBzhGviA1/TUGJLmSVqs8pb9AnvICXEuOHLprYTw=="], + "google-auth-library/gaxios/node-fetch/data-uri-to-buffer": ["data-uri-to-buffer@4.0.1", "", {}, "sha512-0R9ikRb668HB7QDxT1vkpuUBtqc53YyAwMwGeUFKRojY/NWKvdZ+9UYtRfGmhqNbRkTSVpMbmyhXipFFv2cb/A=="], "prebuild-install/tar-fs/tar-stream/readable-stream": ["readable-stream@3.6.2", "", { "dependencies": { "inherits": "^2.0.3", "string_decoder": "^1.1.1", "util-deprecate": "^1.0.1" } }, "sha512-9u/sniCrY3D5WdsERHzHE4G2YCXqoG5FTHUiCC4SIbr6XcLZBY05ya9EKjYek9O5xOAwjGq+1JdGBAS7Q9ScoA=="], @@ -5137,10 +5164,6 @@ "create-agentuity/@agentuity/cli/@agentuity/coder-tui/@mariozechner/pi-tui/chalk": ["chalk@5.6.2", "", {}, "sha512-7NzBL0rN6fMUW+f7A6Io4h40qQlG+xGmtMxfbnH/K7TAtt8JQWVQK+6g0UXKMeVJoyV5EkkNsErQ8pVD3bLHbA=="], - "gcp-metadata/gaxios/node-fetch/whatwg-url/tr46": ["tr46@0.0.3", "", {}, "sha512-N3WMsuqV66lT30CrXNbEjx4GEwlow3v6rr4mCcv6prnfwhS01rkgyFdjPNBYd9br7LpXV1+Emh01fHnq2Gdgrw=="], - - "gcp-metadata/gaxios/node-fetch/whatwg-url/webidl-conversions": ["webidl-conversions@3.0.1", "", {}, "sha512-2JAn3z8AR6rjK8Sm8orRC0h/bcl/DqL7tRPdGZ4I1CjdF+EaMLmYxBHyXuKL849eucPFhvBoxMsflfOb8kxaeQ=="], - "archiver-utils/glob/jackspeak/@isaacs/cliui/string-width/emoji-regex": ["emoji-regex@9.2.2", "", {}, "sha512-L18DaJsXSUk2+42pv8mLs5jJT2hqFkFE4j21wOmgbUqsZ2hL72NsUU785g9RXgo3s0ZNgVl42TiHp3ZtOv/Vyg=="], "archiver-utils/glob/jackspeak/@isaacs/cliui/strip-ansi/ansi-regex": ["ansi-regex@6.2.2", "", {}, "sha512-Bq3SmSpyFHaWjPk8If9yc6svM8c56dB5BAtW4Qbw5jHTwwXXcTLoRMkpDJp6VL0XzlWaCHTXrkFURMYmD0sLqg=="], diff --git a/examples/README.md b/examples/README.md index 7cf6dd6ee..9af19cc82 100644 --- a/examples/README.md +++ b/examples/README.md @@ -16,6 +16,16 @@ AI SDK integration with streaming LLM responses using Vercel AI SDK. - Streaming AI responses - Type-safe LLM interactions +#### [services-aigateway](./services-aigateway/) + +AI Gateway service usage with the standalone TypeScript API. + +**Demonstrates:** + +- Model discovery +- Model metadata filtering +- Routed LLM completions + #### [streaming](./streaming/) Streaming responses using ReadableStream for chunked data delivery. @@ -147,7 +157,7 @@ When adding a new example: | Category | Examples | Description | | ------------------ | ------------------------ | ------------------------------------------- | -| **AI & Streaming** | ai-sdk, streaming | AI integration and data streaming | +| **AI & Streaming** | ai-sdk, services-aigateway, streaming | AI integration and data streaming | | **Real-Time** | websocket, sse | Bidirectional and server-push communication | | **Agent Patterns** | events, evals, lifecycle | Agent features and lifecycle | | **Storage** | services-keyvalue | Data persistence patterns | @@ -163,8 +173,9 @@ When adding a new example: **Advanced patterns?** Check out: 1. [ai-sdk](./ai-sdk/) - AI integration -2. [websocket](./websocket/) - Real-time communication -3. [evals](./evals/) - Quality testing +2. [services-aigateway](./services-aigateway/) - AI Gateway model discovery and completions +3. [websocket](./websocket/) - Real-time communication +4. [evals](./evals/) - Quality testing ## Resources diff --git a/examples/services-aigateway/README.md b/examples/services-aigateway/README.md new file mode 100644 index 000000000..6335a4ef3 --- /dev/null +++ b/examples/services-aigateway/README.md @@ -0,0 +1,76 @@ +# AI Gateway Service Example + +This example demonstrates how to use the standalone `@agentuity/aigateway` TypeScript API from an Agentuity agent. + +## Features Demonstrated + +- **Model discovery** - List AI Gateway models grouped by provider +- **Filtering** - Filter models by provider, input modality, and reasoning support +- **Completions** - Run routed LLM completions through AI Gateway +- **Standalone client** - Use `AIGatewayClient` inside an Agentuity runtime app + +## Running the Example + +```bash +cd examples/services-aigateway +bun install +bun run dev +``` + +## Testing the AI Gateway API Directly + +```bash +# List all models +curl https://aigateway-usc.agentuity.cloud/models + +# List OpenAI models +curl https://aigateway-usc.agentuity.cloud/models/openai + +# Run a completion +curl https://aigateway-usc.agentuity.cloud/ \ + -H "Authorization: Bearer $AGENTUITY_AIGATEWAY_KEY" \ + -H "x-agentuity-orgid: $AGENTUITY_CLOUD_ORG_ID" \ + --json '{"model":"openai/gpt-4.1-mini","messages":[{"role":"user","content":"Say hello in one sentence."}]}' +``` + +## Key Concepts + +### Client Setup + +```typescript +import { AIGatewayClient } from '@agentuity/aigateway'; + +const client = new AIGatewayClient(); +``` + +The client uses standard Agentuity environment variables: + +- `AGENTUITY_AIGATEWAY_KEY` +- `AGENTUITY_SDK_KEY` +- `AGENTUITY_REGION` +- `AGENTUITY_AIGATEWAY_URL` + +### Model Discovery + +```typescript +const catalog = await client.listModels(); +for (const [provider, models] of Object.entries(catalog)) { + console.log(provider, models.map((model) => model.id)); +} +``` + +### Completion + +```typescript +const completion = await client.complete({ + model: 'openai/gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say hello' }], +}); +``` + +## Common Use Cases + +- **Provider-agnostic LLM calls** - Route completion requests through AI Gateway +- **Model picker UIs** - Populate dropdowns from live model metadata +- **Capability filtering** - Select models by modality or reasoning support +- **Centralized billing and auth** - Use Agentuity credentials instead of provider-specific keys diff --git a/examples/services-aigateway/app.ts b/examples/services-aigateway/app.ts new file mode 100644 index 000000000..f2d5ea0ba --- /dev/null +++ b/examples/services-aigateway/app.ts @@ -0,0 +1,6 @@ +import { createApp } from '@agentuity/runtime'; +import aigateway from './src/agent/aigateway/agent'; + +export default await createApp({ + agents: [aigateway], +}); diff --git a/examples/services-aigateway/package.json b/examples/services-aigateway/package.json new file mode 100644 index 000000000..044c05bb2 --- /dev/null +++ b/examples/services-aigateway/package.json @@ -0,0 +1,18 @@ +{ + "name": "example-services-aigateway", + "version": "0.0.1", + "license": "Apache-2.0", + "private": true, + "type": "module", + "scripts": { + "build": "agentuity build --dir .", + "dev": "agentuity dev --dir ." + }, + "dependencies": { + "@agentuity/aigateway": "workspace:*", + "@agentuity/cli": "workspace:*", + "@agentuity/core": "workspace:*", + "@agentuity/runtime": "workspace:*", + "@agentuity/schema": "workspace:*" + } +} diff --git a/examples/services-aigateway/src/agent/aigateway/agent.ts b/examples/services-aigateway/src/agent/aigateway/agent.ts new file mode 100644 index 000000000..8b78ca585 --- /dev/null +++ b/examples/services-aigateway/src/agent/aigateway/agent.ts @@ -0,0 +1,107 @@ +/** + * AI Gateway Example + * + * Demonstrates how to use the standalone @agentuity/aigateway TypeScript API + * from an Agentuity agent. + */ + +import { AIGatewayClient } from '@agentuity/aigateway'; +import { createAgent } from '@agentuity/runtime'; +import { s } from '@agentuity/schema'; + +const inputSchema = s.union([ + s.object({ + operation: s.literal('models'), + provider: s.string().optional(), + input: s.string().optional(), + reasoning: s.boolean().optional(), + }), + s.object({ + operation: s.literal('complete'), + model: s.string(), + prompt: s.string(), + system: s.string().optional(), + temperature: s.number().optional(), + maxTokens: s.number().optional(), + }), +]); + +function getCompletionText(response: unknown): string { + const choices = (response as { choices?: unknown }).choices; + const first = + Array.isArray(choices) && choices.length > 0 + ? (choices[0] as { message?: { content?: unknown }; text?: unknown }) + : undefined; + const content = + first?.message?.content ?? first?.text ?? (response as { content?: unknown }).content; + if (typeof content === 'string') { + return content; + } + if (Array.isArray(content)) { + return content + .map((part) => { + if (typeof part === 'string') return part; + if (part && typeof part === 'object' && 'text' in part) { + const text = (part as { text?: unknown }).text; + return typeof text === 'string' ? text : ''; + } + return ''; + }) + .join(''); + } + return ''; +} + +export default createAgent('aigateway', { + description: 'Example agent demonstrating AI Gateway model discovery and completions', + schema: { + input: inputSchema, + output: s.any(), + }, + handler: async (ctx, input) => { + const client = new AIGatewayClient({ logger: ctx.logger }); + + switch (input.operation) { + case 'models': { + const catalog = await client.listModels(); + const models = Object.entries(catalog) + .filter(([provider]) => !input.provider || provider === input.provider) + .flatMap(([provider, providerModels]) => + providerModels + .filter( + (model) => !input.input || model.input_modalities?.includes(input.input) + ) + .filter((model) => !input.reasoning || model.reasoning) + .map((model) => ({ + provider, + id: model.id, + name: model.name, + api: model.api, + contextWindow: model.context_window, + maxOutputTokens: model.max_output_tokens, + reasoning: model.reasoning, + inputModalities: model.input_modalities, + outputModalities: model.output_modalities, + })) + ); + return { models, count: models.length }; + } + + case 'complete': { + const response = await client.complete({ + model: input.model, + messages: [ + ...(input.system ? [{ role: 'system' as const, content: input.system }] : []), + { role: 'user' as const, content: input.prompt }, + ], + temperature: input.temperature, + max_tokens: input.maxTokens, + }); + return { + text: getCompletionText(response), + response, + }; + } + } + }, +}); diff --git a/package.json b/package.json index fbef8b775..d6239f6f6 100644 --- a/package.json +++ b/package.json @@ -9,8 +9,8 @@ ], "scripts": { "prepare": "husky", - "build": "bunx tsc --build && bun run --filter='./packages/core' build && bun run --filter='./packages/schema' build && bun run --filter='./packages/frontend' build && bun run --filter='./packages/server' build && bun run --filter='./packages/react' build && bun run --filter='./packages/postgres' build && bun run --filter='./packages/drizzle' build && bun run --filter='./packages/auth' build && bun run --filter='./packages/evals' build && bun run --filter='./packages/workbench' build && bun run --filter='./packages/runtime' build && bun packages/frontend/scripts/build-beacon.ts && bun run --filter='./packages/cli' build && bun run --filter='./packages/opencode' build && bun run --filter='./apps/testing/integration-suite' build && bun run --filter='./apps/testing/cloud-deployment' build && bun run --filter='./apps/testing/e2e-web' build && bun run --filter='./apps/testing/svelte-web' build", - "build:packages": "bunx tsc --build && bun run --filter='./packages/core' build && bun run --filter='./packages/schema' build && bun run --filter='./packages/frontend' build && bun run --filter='./packages/server' build && bun run --filter='./packages/react' build && bun run --filter='./packages/postgres' build && bun run --filter='./packages/drizzle' build && bun run --filter='./packages/auth' build && bun run --filter='./packages/evals' build && bun run --filter='./packages/workbench' build && bun run --filter='./packages/runtime' build && bun packages/frontend/scripts/build-beacon.ts && bun run --filter='./packages/cli' build && bun run --filter='./packages/opencode' build", + "build": "bunx tsc --build && bun run --filter='./packages/core' build && bun run --filter='./packages/schema' build && bun run --filter='./packages/frontend' build && bun run --filter='./packages/server' build && bun run --filter='./packages/react' build && bun run --filter='./packages/postgres' build && bun run --filter='./packages/drizzle' build && bun run --filter='./packages/auth' build && bun run --filter='./packages/aigateway' build && bun run --filter='./packages/evals' build && bun run --filter='./packages/workbench' build && bun run --filter='./packages/runtime' build && bun packages/frontend/scripts/build-beacon.ts && bun run --filter='./packages/cli' build && bun run --filter='./packages/opencode' build && bun run --filter='./apps/testing/integration-suite' build && bun run --filter='./apps/testing/cloud-deployment' build && bun run --filter='./apps/testing/e2e-web' build && bun run --filter='./apps/testing/svelte-web' build", + "build:packages": "bunx tsc --build && bun run --filter='./packages/core' build && bun run --filter='./packages/schema' build && bun run --filter='./packages/frontend' build && bun run --filter='./packages/server' build && bun run --filter='./packages/react' build && bun run --filter='./packages/postgres' build && bun run --filter='./packages/drizzle' build && bun run --filter='./packages/auth' build && bun run --filter='./packages/aigateway' build && bun run --filter='./packages/evals' build && bun run --filter='./packages/workbench' build && bun run --filter='./packages/runtime' build && bun packages/frontend/scripts/build-beacon.ts && bun run --filter='./packages/cli' build && bun run --filter='./packages/opencode' build", "dev:workbench": "concurrently \"cd packages/workbench && bun run dev:app\" \"cd apps/testing/integration-suite && bun run dev\" --names \"workbench,integration-suite\" --prefix-colors \"blue,green\"", "test": "bun test:packages && cd packages/cli && bun run test && cd ../.. && bun test:templates && bun test:pkginstall && bun test:create", "test:packages": "cd packages/core && bun test && cd ../schema && bun test && cd ../frontend && bun test && cd ../server && bun test && cd ../react && bun test && cd ../postgres && bun test && cd ../drizzle && bun test && cd ../auth && bun test && cd ../runtime && bun test --max-concurrency=1 && cd ../opencode && bun test", diff --git a/packages/aigateway/README.md b/packages/aigateway/README.md new file mode 100644 index 000000000..d74029f88 --- /dev/null +++ b/packages/aigateway/README.md @@ -0,0 +1,50 @@ +# @agentuity/aigateway + +A standalone package for the Agentuity AI Gateway service. + +## Installation + +```bash +npm install @agentuity/aigateway +``` + +## Quick Start + +```typescript +import { AIGatewayClient } from '@agentuity/aigateway'; + +const client = new AIGatewayClient(); + +const models = await client.listModels(); +console.log(Object.keys(models)); + +const completion = await client.complete({ + model: 'openai/gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say hello' }], +}); + +console.log(completion.choices?.[0]); +``` + +## Configuration + +```typescript +const client = new AIGatewayClient({ + apiKey: 'your-api-key', + orgId: 'your-org-id', + url: 'https://aigateway-usc.agentuity.cloud', +}); +``` + +### Environment Variables + +| Variable | Description | Default | +|----------|-------------|---------| +| `AGENTUITY_AIGATEWAY_KEY` | AI Gateway API key override | Optional | +| `AGENTUITY_SDK_KEY` | API key for authentication | Required | +| `AGENTUITY_REGION` | Region for API endpoints | `usc` | +| `AGENTUITY_AIGATEWAY_URL` | Override AI Gateway API URL | Auto-detected | + +## License + +Apache-2.0 diff --git a/packages/aigateway/package.json b/packages/aigateway/package.json new file mode 100644 index 000000000..121dd26e3 --- /dev/null +++ b/packages/aigateway/package.json @@ -0,0 +1,42 @@ +{ + "name": "@agentuity/aigateway", + "version": "2.0.14", + "license": "Apache-2.0", + "author": "Agentuity employees and contributors", + "type": "module", + "main": "./dist/index.js", + "types": "./dist/index.d.ts", + "files": [ + "AGENTS.md", + "README.md", + "src", + "dist" + ], + "exports": { + ".": { + "import": "./dist/index.js", + "types": "./dist/index.d.ts" + } + }, + "scripts": { + "clean": "rm -rf dist tsconfig.tsbuildinfo", + "build": "bunx tsc --build --force", + "typecheck": "bunx tsc --noEmit", + "prepublishOnly": "bun run clean && bun run build" + }, + "dependencies": { + "@agentuity/core": "workspace:*", + "@agentuity/server": "workspace:*", + "zod": "^4.3.5" + }, + "devDependencies": { + "@types/bun": "latest", + "@types/node": "^22.0.0", + "bun-types": "latest", + "typescript": "^5.9.0" + }, + "publishConfig": { + "access": "public" + }, + "sideEffects": false +} diff --git a/packages/aigateway/src/index.ts b/packages/aigateway/src/index.ts new file mode 100644 index 000000000..a634d0146 --- /dev/null +++ b/packages/aigateway/src/index.ts @@ -0,0 +1,79 @@ +export { + AIGatewayService, + type AIGatewayChatCompletion, + type AIGatewayChatCompletionParams, + type AIGatewayChatMessage, + type AIGatewayModel, + type AIGatewayModelProvider, + type AIGatewayModels, + type AIGatewayModelsResponse, + type AIGatewayPricing, + AIGatewayChatCompletionParamsSchema, + AIGatewayChatCompletionSchema, + AIGatewayChatMessageSchema, + AIGatewayModelProviderSchema, + AIGatewayModelSchema, + AIGatewayModelsResponseSchema, + AIGatewayModelsSchema, + AIGatewayPricingSchema, +} from '@agentuity/core/aigateway'; + +import { + AIGatewayService, + type AIGatewayChatCompletion, + type AIGatewayChatCompletionParams, + type AIGatewayModels, +} from '@agentuity/core/aigateway'; +import { createMinimalLogger, getEnv } from '@agentuity/core'; +import { getServiceUrls } from '@agentuity/core/config'; +import { buildClientHeaders, createServerFetchAdapter, type Logger } from '@agentuity/server'; +import { z } from 'zod'; + +const isLogger = (val: unknown): val is Logger => + typeof val === 'object' && + val !== null && + ['info', 'warn', 'error', 'debug', 'trace'].every( + (m) => typeof (val as Record)[m] === 'function' + ); + +export const AIGatewayClientOptionsSchema = z.object({ + apiKey: z.string().optional().describe('API key for authentication'), + url: z.string().optional().describe('Base URL for the AI Gateway API'), + orgId: z.string().optional().describe('Organization ID for multi-tenant operations'), + logger: z.custom(isLogger).optional().describe('Custom logger instance'), +}); + +export type AIGatewayClientOptions = z.infer; + +export class AIGatewayClient { + readonly #service: AIGatewayService; + + constructor(options: AIGatewayClientOptions = {}) { + const validatedOptions = AIGatewayClientOptionsSchema.parse(options); + const apiKey = + validatedOptions.apiKey || + getEnv('AGENTUITY_AIGATEWAY_KEY') || + getEnv('AGENTUITY_SDK_KEY') || + getEnv('AGENTUITY_CLI_KEY'); + const region = getEnv('AGENTUITY_REGION') ?? 'usc'; + const serviceUrls = getServiceUrls(region); + const url = + validatedOptions.url || getEnv('AGENTUITY_AIGATEWAY_URL') || serviceUrls.aigateway; + const logger = validatedOptions.logger ?? createMinimalLogger(); + const headers = buildClientHeaders({ + apiKey, + orgId: validatedOptions.orgId, + }); + + const adapter = createServerFetchAdapter({ headers }, logger); + this.#service = new AIGatewayService(url, adapter); + } + + async listModels(): Promise { + return this.#service.listModels(); + } + + async complete(params: AIGatewayChatCompletionParams): Promise { + return this.#service.complete(params); + } +} diff --git a/packages/aigateway/tsconfig.json b/packages/aigateway/tsconfig.json new file mode 100644 index 000000000..39771b3c1 --- /dev/null +++ b/packages/aigateway/tsconfig.json @@ -0,0 +1,11 @@ +{ + "extends": "../../tsconfig.base.json", + "compilerOptions": { + "composite": true, + "outDir": "./dist", + "rootDir": "./src" + }, + "include": ["src/**/*"], + "exclude": ["node_modules", "dist"], + "references": [{ "path": "../core" }, { "path": "../server" }] +} diff --git a/packages/cli/src/agent-detection.ts b/packages/cli/src/agent-detection.ts index ed0f565ba..73e4215a5 100644 --- a/packages/cli/src/agent-detection.ts +++ b/packages/cli/src/agent-detection.ts @@ -33,6 +33,7 @@ export const KNOWN_AGENTS: [string, string][] = [ ['amp', 'amp'], ['warp', 'warp'], ['pi', 'pi'], + ['coder', 'coder'], // TODO: VSCode Agent Mode detection - need to find a reliable way to detect // when VSCode's built-in agent (Copilot Chat) is running commands vs just // running in VSCode's integrated terminal. May need env var detection. @@ -342,14 +343,10 @@ let cachedResult: string | undefined | null = null; /** * Check if a basename matches a known agent process name. - * Short tokens (≤2 chars) require an exact match to avoid false positives - * (e.g., 'pi' matching 'pip', 'spin'). Longer tokens use substring matching. + * Match exact executable names or names split on non-alphanumeric boundaries. */ function matchesProcessName(basename: string, processName: string): boolean { - if (processName.length <= 2) { - return basename === processName; - } - return basename.includes(processName); + return basename === processName || basename.split(/[^a-z0-9]+/i).includes(processName); } /** diff --git a/packages/cli/src/ai-help.ts b/packages/cli/src/ai-help.ts index f1210886f..1f8810de3 100644 --- a/packages/cli/src/ai-help.ts +++ b/packages/cli/src/ai-help.ts @@ -115,7 +115,7 @@ function buildWhenToUse(): string { Use this CLI when the user asks to: - Create, build, or deploy AI agents - Manage Agentuity projects and organizations -- Access cloud services (KV, Vector, Postgres, Storage, Sandboxes) +- Access cloud services (KV, Vector, AI Gateway, Postgres, Storage, Sandboxes) - Debug or troubleshoot agent deployments - Run agents locally in development mode - Manage environment variables and secrets @@ -228,15 +228,15 @@ function collectQuickReferenceCommands(schema: CLISchema): QuickCommand[] { { path: 'cloud kv get', label: 'KV get', priority: 9 }, { path: 'cloud kv set', label: 'KV set', priority: 10 }, { path: 'cloud vector search', label: 'Vector search', priority: 11 }, - { path: 'cloud db list', label: 'List databases', priority: 12 }, - { path: 'cloud db exec', label: 'Execute SQL', priority: 13 }, - { path: 'env list', label: 'List env vars', priority: 14 }, - { path: 'env set', label: 'Set env var', priority: 15 }, - { path: 'cloud deployment list', label: 'List deployments', priority: 16 }, - { path: 'cloud deployment rollback', label: 'Rollback deployment', priority: 17 }, - { path: 'project list', label: 'List projects', priority: 18 }, - { path: 'auth whoami', label: 'Show current user', priority: 19 }, - { path: 'version', label: 'Show version', priority: 20 }, + { path: 'cloud aigateway models', label: 'List AI models', priority: 12 }, + { path: 'cloud aigateway complete', label: 'AI completion', priority: 13 }, + { path: 'cloud db list', label: 'List databases', priority: 14 }, + { path: 'cloud db exec', label: 'Execute SQL', priority: 15 }, + { path: 'env list', label: 'List env vars', priority: 16 }, + { path: 'env set', label: 'Set env var', priority: 17 }, + { path: 'cloud deployment list', label: 'List deployments', priority: 18 }, + { path: 'cloud deployment rollback', label: 'Rollback deployment', priority: 19 }, + { path: 'project list', label: 'List projects', priority: 20 }, ]; // Build command signatures from schema diff --git a/packages/cli/src/cmd/ai/capabilities/show.ts b/packages/cli/src/cmd/ai/capabilities/show.ts index 7e4cafbbe..0804118f4 100644 --- a/packages/cli/src/cmd/ai/capabilities/show.ts +++ b/packages/cli/src/cmd/ai/capabilities/show.ts @@ -128,6 +128,12 @@ export const showSubcommand = createSubcommand({ requiresAuth: true, requiresProject: true, }, + { + id: 'aigateway', + name: 'AI Gateway', + description: 'List supported AI models and run LLM completions', + commands: ['cloud aigateway models', 'cloud aigateway complete'], + }, { id: 'databases', name: 'Cloud Databases', diff --git a/packages/cli/src/cmd/ai/intro.ts b/packages/cli/src/cmd/ai/intro.ts index 7433badd2..4bdd9cb36 100644 --- a/packages/cli/src/cmd/ai/intro.ts +++ b/packages/cli/src/cmd/ai/intro.ts @@ -55,6 +55,7 @@ ${getCommand('cloud deployment logs')} # View deployment logs \`\`\`bash ${getCommand('cloud kv')} # Key-value storage operations ${getCommand('cloud vector')} # Vector database operations +${getCommand('cloud aigateway')} # AI Gateway model and completion operations ${getCommand('cloud storage')} # Object storage operations ${getCommand('env set KEY value')} # Set environment variables ${getCommand('env set KEY value --secret')} # Set secrets (encrypted) diff --git a/packages/cli/src/cmd/cloud/aigateway/complete.ts b/packages/cli/src/cmd/cloud/aigateway/complete.ts new file mode 100644 index 000000000..505f2507f --- /dev/null +++ b/packages/cli/src/cmd/cloud/aigateway/complete.ts @@ -0,0 +1,461 @@ +import { z } from 'zod'; +import { StructuredError, type AIGatewayModels, type AIGatewayService } from '@agentuity/core'; +import { createCommand } from '../../../types'; +import { getCommand } from '../../../command-prefix'; +import { getExecutingAgent } from '../../../agent-detection'; +import { createAIGatewayService, getAIGatewayUrl, getCompletionText } from './util'; +import { getCachedAIGatewayModels, setCachedAIGatewayModels } from './model-cache'; + +const CompletionResponseSchema = z.object({ + text: z.string(), + response: z.unknown(), + cost: z.unknown().optional(), +}); + +const defaultModel = 'openai/gpt-4o-mini'; +const PromptRequiredError = StructuredError( + 'AIGatewayPromptRequired', + 'Prompt is required. Pass it as an argument, use --prompt, use --file, or pipe it through stdin.' +); +const PromptFileNotFoundError = StructuredError('AIGatewayPromptFileNotFound')<{ + filename: string; +}>(); + +function isAgentOutputMode(): boolean { + return Boolean(getExecutingAgent()) && process.env.AGENTUITY_AIGATEWAY_AGENT_OUTPUT !== 'false'; +} + +async function readPromptFromStdin(): Promise { + if (process.stdin.isTTY) { + return undefined; + } + const text = await Bun.stdin.text(); + const trimmed = text.trim(); + return trimmed.length > 0 ? trimmed : undefined; +} + +async function readPromptFromFile(filename?: string): Promise { + if (!filename) { + return undefined; + } + const file = Bun.file(filename); + if (!(await file.exists())) { + throw new PromptFileNotFoundError({ + message: `Prompt file not found: ${filename}`, + filename, + }); + } + const text = await file.text(); + const trimmed = text.trim(); + return trimmed.length > 0 ? trimmed : undefined; +} + +export function combinePromptInput(opts: { + explicitPrompt?: string; + stdinPrompt?: string; + stdinMode?: 'append' | 'replace'; +}): string | undefined { + if (!opts.stdinPrompt) { + return opts.explicitPrompt; + } + if (!opts.explicitPrompt || opts.stdinMode === 'replace') { + return opts.stdinPrompt; + } + if (!opts.stdinMode || opts.stdinMode === 'append') { + return `${opts.explicitPrompt}\n\n${opts.stdinPrompt}`; + } + return opts.explicitPrompt; +} + +function getUsageText(response: unknown): string | undefined { + if (!response || typeof response !== 'object') { + return undefined; + } + const usage = (response as { usage?: unknown }).usage; + if (!usage || typeof usage !== 'object') { + return undefined; + } + const input = + (usage as { prompt_tokens?: unknown; input_tokens?: unknown }).prompt_tokens ?? + (usage as { input_tokens?: unknown }).input_tokens; + const output = + (usage as { completion_tokens?: unknown; output_tokens?: unknown }).completion_tokens ?? + (usage as { output_tokens?: unknown }).output_tokens; + const total = (usage as { total_tokens?: unknown }).total_tokens; + const parts = [ + typeof input === 'number' ? `input=${input}` : undefined, + typeof output === 'number' ? `output=${output}` : undefined, + typeof total === 'number' ? `total=${total}` : undefined, + ].filter(Boolean); + return parts.length > 0 ? `Usage: ${parts.join(' ')}` : undefined; +} + +function getCostInfo(response: unknown): unknown | undefined { + if (!response || typeof response !== 'object') { + return undefined; + } + const agentuity = (response as { agentuity?: unknown }).agentuity; + if (!agentuity || typeof agentuity !== 'object') { + return undefined; + } + return (agentuity as { cost?: unknown }).cost; +} + +function getCostText(response: unknown): string | undefined { + const cost = getCostInfo(response); + if (!cost || typeof cost !== 'object') { + return undefined; + } + const total = (cost as { total?: unknown }).total; + const promptTokens = (cost as { promptTokens?: unknown }).promptTokens; + const completionTokens = (cost as { completionTokens?: unknown }).completionTokens; + const parts = [ + typeof total === 'number' ? `total=$${total.toFixed(6)}` : undefined, + typeof promptTokens === 'number' ? `prompt=${promptTokens}` : undefined, + typeof completionTokens === 'number' ? `completion=${completionTokens}` : undefined, + ].filter(Boolean); + return parts.length > 0 ? `Cost: ${parts.join(' ')}` : undefined; +} + +type CompletionModelInfo = { + id: string; + api?: string; + provider?: string; +}; + +function matchesModel(provider: string, candidateId: string, model: string): boolean { + return candidateId === model || `${provider}/${candidateId}` === model; +} + +async function getCompletionModelInfo( + model: string, + models: AIGatewayModels +): Promise { + for (const [provider, providerModels] of Object.entries(models)) { + const match = providerModels.find((candidate) => matchesModel(provider, candidate.id, model)); + if (match) { + return { id: match.id, api: match.api, provider }; + } + } + return undefined; +} + +async function loadModelsForCompletion(opts: { + service: AIGatewayService; + profile: string; + cacheKey: string; + refresh?: boolean; +}): Promise { + if (!opts.refresh) { + const cached = await getCachedAIGatewayModels(opts.profile, opts.cacheKey); + if (cached) { + return cached; + } + } + const models = await opts.service.listModels(); + await setCachedAIGatewayModels(opts.profile, opts.cacheKey, models); + return models; +} + +function buildCompletionRequest(opts: { + model: string; + prompt: string; + system?: string; + temperature?: number; + maxTokens?: number; + stream?: boolean; +}) { + const common = { + model: opts.model, + temperature: opts.temperature, + max_tokens: opts.maxTokens, + ...(opts.stream ? { stream: true } : {}), + }; + return { + ...common, + messages: [ + ...(opts.system ? [{ role: 'system' as const, content: opts.system }] : []), + { role: 'user' as const, content: opts.prompt }, + ], + }; +} + +async function resolvePrompt(opts: { + optionPrompt?: string; + argPrompt?: string; + file?: string; + stdinMode?: 'append' | 'replace'; +}): Promise { + const explicitPrompt = + opts.optionPrompt ?? opts.argPrompt ?? (await readPromptFromFile(opts.file)); + const stdinPrompt = await readPromptFromStdin(); + return combinePromptInput({ explicitPrompt, stdinPrompt, stdinMode: opts.stdinMode }); +} + +function getStreamDeltaText(payload: unknown): string { + if (!payload || typeof payload !== 'object') { + return ''; + } + const choices = (payload as { choices?: unknown }).choices; + if (!Array.isArray(choices)) { + return ''; + } + return choices + .map((choice) => { + if (!choice || typeof choice !== 'object') { + return ''; + } + const delta = (choice as { delta?: { content?: unknown } }).delta; + if (typeof delta?.content === 'string') { + return delta.content; + } + const text = (choice as { text?: unknown }).text; + return typeof text === 'string' ? text : ''; + }) + .join(''); +} + +async function consumeCompletionStream( + stream: ReadableStream, + options: { json?: boolean; raw?: boolean } +): Promise { + const reader = stream.getReader(); + const decoder = new TextDecoder(); + let buffer = ''; + let text = ''; + + const consumeFrame = (frame: string) => { + const dataLines = frame + .split(/\r?\n/) + .filter((line) => line.startsWith('data:')) + .map((line) => line.slice(5).trimStart()); + for (const data of dataLines) { + if (!data || data === '[DONE]') { + continue; + } + try { + const delta = getStreamDeltaText(JSON.parse(data)); + if (delta) { + text += delta; + } + } catch { + // Ignore malformed stream frames and continue consuming the stream. + } + if (options.raw) { + if (!options.json) { + console.log(data); + } + continue; + } + try { + const delta = getStreamDeltaText(JSON.parse(data)); + if (delta) { + if (!options.json) { + process.stdout.write(delta); + } + } + } catch { + // Ignore malformed stream frames and continue consuming the stream. + } + } + }; + + try { + while (true) { + const { done, value } = await reader.read(); + if (done) { + break; + } + buffer += decoder.decode(value, { stream: true }); + const frames = buffer.split(/\r?\n\r?\n/); + buffer = frames.pop() ?? ''; + for (const frame of frames) { + consumeFrame(frame); + } + } + buffer += decoder.decode(); + if (buffer.trim()) { + consumeFrame(buffer); + } + } finally { + reader.releaseLock(); + } + if (!options.json && !options.raw && text) { + process.stdout.write('\n'); + } + return text; +} + +export const completeSubcommand = createCommand({ + name: 'complete', + aliases: ['completion', 'chat'], + description: 'Run an AI Gateway completion', + tags: ['write', 'slow', 'requires-auth', 'uses-stdin'], + requires: { auth: true }, + optional: { project: true, region: true }, + examples: [ + { + command: getCommand('cloud aigateway complete --model openai/gpt-4.1-mini "Hello"'), + description: 'Run a completion', + }, + { + command: `echo "Hello" | ${getCommand('cloud aigateway complete --model openai/gpt-4.1-mini')}`, + description: 'Read the prompt from stdin', + }, + { + command: getCommand( + 'cloud aigateway complete --model openai/gpt-4.1-mini --file prompt.txt' + ), + description: 'Read the prompt from a file', + }, + { + command: getCommand( + 'cloud aigateway complete --model openai/gpt-4.1-mini --stream "Hello"' + ), + description: 'Stream token output as it arrives', + }, + ], + schema: { + args: z.object({ + prompt: z.string().optional().describe('prompt text'), + }), + options: z.object({ + model: z.string().min(1).optional().describe('model id'), + prompt: z.string().optional().describe('prompt text'), + file: z.string().optional().describe('read prompt text from a file'), + system: z.string().optional().describe('optional system message'), + systemFile: z.string().optional().describe('read the system message from a file'), + refreshModels: z + .boolean() + .optional() + .describe('refresh the cached AI Gateway model catalog before choosing request format'), + temperature: z.number().optional().describe('sampling temperature'), + maxTokens: z.number().optional().describe('maximum output tokens'), + stream: z.boolean().optional().describe('stream token output as it arrives'), + save: z.string().optional().describe('write assistant text to a file'), + format: z + .enum(['text', 'json', 'raw']) + .optional() + .describe('output format for non-json mode'), + stdinMode: z + .enum(['append', 'replace']) + .optional() + .describe('how to combine stdin with prompt text'), + usage: z.boolean().optional().describe('print usage details when available'), + cost: z.boolean().optional().describe('print AI Gateway cost details when available'), + raw: z.boolean().optional().describe('print the raw completion response'), + }), + response: CompletionResponseSchema, + }, + async handler(ctx) { + const prompt = await resolvePrompt({ + optionPrompt: ctx.opts.prompt, + argPrompt: ctx.args.prompt, + file: ctx.opts.file, + stdinMode: ctx.opts.stdinMode, + }); + if (!prompt) { + throw new PromptRequiredError(); + } + + const service = createAIGatewayService(ctx); + const model = ctx.opts.model ?? process.env.AGENTUITY_AIGATEWAY_MODEL ?? defaultModel; + const system = ctx.opts.system ?? (await readPromptFromFile(ctx.opts.systemFile)); + const profile = ctx.config?.name ?? 'default'; + const cacheKey = getAIGatewayUrl(ctx.region, ctx.config?.overrides); + let models = await loadModelsForCompletion({ + service, + profile, + cacheKey, + refresh: ctx.opts.refreshModels, + }); + let modelInfo = await getCompletionModelInfo(model, models); + if (!modelInfo && !ctx.opts.refreshModels) { + models = await loadModelsForCompletion({ service, profile, cacheKey, refresh: true }); + modelInfo = await getCompletionModelInfo(model, models); + } + const requestModel = modelInfo?.id ?? model; + const request = buildCompletionRequest({ + model: requestModel, + prompt, + system, + temperature: ctx.opts.temperature, + maxTokens: ctx.opts.maxTokens, + }); + const format = ctx.opts.raw + ? 'raw' + : (ctx.opts.format ?? (isAgentOutputMode() ? 'json' : 'text')); + + if (ctx.opts.stream) { + const streamed = await service.streamCompleteWithMetadata({ ...request, stream: true }); + const text = await consumeCompletionStream(streamed.stream, { + json: ctx.options.json || format === 'json', + raw: format === 'raw', + }); + const metadata = await streamed.metadata; + const cost = metadata.cost; + if (ctx.opts.save) { + await Bun.write(ctx.opts.save, text); + } + if (!ctx.options.json && format === 'json') { + console.log( + JSON.stringify( + { text, cost, response: { stream: true, model: requestModel } }, + null, + 2 + ) + ); + } + if (!ctx.options.json && ctx.opts.cost) { + const costText = getCostText({ agentuity: metadata }); + if (costText) { + console.error(costText); + } + } + return { text, response: { stream: true, model: requestModel }, cost }; + } + + const response = await service.complete(request); + const text = getCompletionText(response); + const cost = getCostInfo(response); + if (ctx.opts.save) { + await Bun.write(ctx.opts.save, text); + } + + if (!ctx.options.json) { + if (format === 'raw') { + console.log(JSON.stringify(response, null, 2)); + } else if (format === 'json') { + console.log( + JSON.stringify( + { + text, + model: requestModel, + usage: (response as { usage?: unknown }).usage, + cost, + response, + }, + null, + 2 + ) + ); + } else { + console.log(text); + } + if (ctx.opts.usage) { + const usage = getUsageText(response); + if (usage) { + console.error(usage); + } + } + if (ctx.opts.cost) { + const costText = getCostText(response); + if (costText) { + console.error(costText); + } + } + } + + return { text, response, cost }; + }, +}); diff --git a/packages/cli/src/cmd/cloud/aigateway/index.ts b/packages/cli/src/cmd/cloud/aigateway/index.ts new file mode 100644 index 000000000..0370acf08 --- /dev/null +++ b/packages/cli/src/cmd/cloud/aigateway/index.ts @@ -0,0 +1,21 @@ +import { createCommand } from '../../../types'; +import { getCommand } from '../../../command-prefix'; +import { completeSubcommand } from './complete'; +import { modelsSubcommand } from './models'; + +export const aigatewayCommand = createCommand({ + name: 'aigateway', + aliases: ['ai-gateway', 'ai'], + description: 'Use the Agentuity AI Gateway', + tags: ['slow'], + examples: [ + { command: getCommand('cloud aigateway models'), description: 'List supported models' }, + { + command: getCommand('cloud aigateway complete --model openai/gpt-4.1-mini "Hello"'), + description: 'Run a completion', + }, + ], + subcommands: [modelsSubcommand, completeSubcommand], +}); + +export default aigatewayCommand; diff --git a/packages/cli/src/cmd/cloud/aigateway/model-cache.ts b/packages/cli/src/cmd/cloud/aigateway/model-cache.ts new file mode 100644 index 000000000..41c4a1053 --- /dev/null +++ b/packages/cli/src/cmd/cloud/aigateway/model-cache.ts @@ -0,0 +1,89 @@ +import { Database } from 'bun:sqlite'; +import { mkdir } from 'node:fs/promises'; +import { join } from 'node:path'; +import { AIGatewayModelsSchema, type AIGatewayModels } from '@agentuity/core'; +import { getDefaultConfigDir } from '../../../config'; + +const TTL_MS = 6 * 60 * 60 * 1000; + +let db: Database | null = null; + +async function getDatabase(): Promise { + if (db) { + return db; + } + + const configDir = getDefaultConfigDir(); + await mkdir(configDir, { recursive: true }); + + db = new Database(join(configDir, 'resource.db')); + db.run('PRAGMA journal_mode = WAL'); + db.run('PRAGMA busy_timeout = 5000'); + db.run('PRAGMA synchronous = NORMAL'); + db.run(` + CREATE TABLE IF NOT EXISTS aigateway_model_cache ( + profile TEXT NOT NULL, + cache_key TEXT NOT NULL, + models_json TEXT NOT NULL, + cached_at INTEGER NOT NULL, + PRIMARY KEY (profile, cache_key) + ) + `); + db.run(` + CREATE INDEX IF NOT EXISTS idx_aigateway_model_cache_cached_at + ON aigateway_model_cache(cached_at) + `); + + return db; +} + +export async function getCachedAIGatewayModels( + profile: string, + cacheKey: string +): Promise { + try { + const database = await getDatabase(); + const cutoff = Date.now() - TTL_MS; + const row = database + .query<{ models_json: string; cached_at: number }, [string, string]>( + 'SELECT models_json, cached_at FROM aigateway_model_cache WHERE profile = ? AND cache_key = ?' + ) + .get(profile, cacheKey); + if (!row) { + return null; + } + if (row.cached_at < cutoff) { + database.run('DELETE FROM aigateway_model_cache WHERE profile = ? AND cache_key = ?', [ + profile, + cacheKey, + ]); + return null; + } + const parsed = AIGatewayModelsSchema.safeParse(JSON.parse(row.models_json)); + return parsed.success ? parsed.data : null; + } catch { + return null; + } +} + +export async function setCachedAIGatewayModels( + profile: string, + cacheKey: string, + models: AIGatewayModels +): Promise { + try { + const database = await getDatabase(); + const cutoff = Date.now() - TTL_MS; + database.run('DELETE FROM aigateway_model_cache WHERE cached_at < ?', [cutoff]); + database.run( + `INSERT INTO aigateway_model_cache (profile, cache_key, models_json, cached_at) + VALUES (?, ?, ?, ?) + ON CONFLICT(profile, cache_key) DO UPDATE SET + models_json = excluded.models_json, + cached_at = excluded.cached_at`, + [profile, cacheKey, JSON.stringify(models), Date.now()] + ); + } catch { + // Non-critical cache failure should never block the CLI. + } +} diff --git a/packages/cli/src/cmd/cloud/aigateway/models.ts b/packages/cli/src/cmd/cloud/aigateway/models.ts new file mode 100644 index 000000000..c160ec4f0 --- /dev/null +++ b/packages/cli/src/cmd/cloud/aigateway/models.ts @@ -0,0 +1,219 @@ +import { z } from 'zod'; +import { createCommand } from '../../../types'; +import * as tui from '../../../tui'; +import { getCommand } from '../../../command-prefix'; +import { getExecutingAgent } from '../../../agent-detection'; +import { createPublicAIGatewayService, getAIGatewayUrl } from './util'; +import { getCachedAIGatewayModels, setCachedAIGatewayModels } from './model-cache'; + +const ModelRowSchema = z.object({ + provider: z.string(), + id: z.string(), + name: z.string(), + api: z.string().optional(), + reasoning: z.boolean().optional(), + contextWindow: z.number().optional(), + maxOutputTokens: z.number().optional(), +}); + +const ModelsResponseSchema = z.object({ + models: z.array(ModelRowSchema), + count: z.number(), + model: ModelRowSchema.nullable().optional(), +}); + +const recommendedModels = [ + { use: 'fast', candidates: ['openai/gpt-4o-mini', 'openai/gpt-4.1-mini'] }, + { use: 'reasoning', candidates: ['openai/gpt-5-mini', 'openai/o4-mini'] }, + { use: 'coding', candidates: ['anthropic/claude-opus-4-7', 'openai/gpt-5-codex'] }, + { use: 'cheap', candidates: ['openai/gpt-4.1-nano', 'openai/gpt-5-nano'] }, +]; + +function isAgentOutputMode(): boolean { + return Boolean(getExecutingAgent()) && process.env.AGENTUITY_AIGATEWAY_AGENT_OUTPUT !== 'false'; +} + +function getRecommendations(rows: z.infer[]) { + const byId = new Map(rows.map((row) => [normalizeModelId(row.id), row])); + return recommendedModels + .map((rec) => { + const model = rec.candidates.map((id) => byId.get(normalizeModelId(id))).find(Boolean); + return model ? { use: rec.use, model: model.id, name: model.name } : undefined; + }) + .filter((row): row is { use: string; model: string; name: string } => Boolean(row)); +} + +function normalizeModelId(id: string): string { + const normalized = id.toLowerCase(); + const parts = normalized.split('/'); + return parts.length > 1 ? (parts.at(-1) ?? normalized) : normalized; +} + +function matchesProviderFilter( + provider: string, + modelId: string, + providerFilter?: string +): boolean { + if (!providerFilter) { + return true; + } + return provider === providerFilter || modelId.startsWith(`${providerFilter}/`); +} + +function matchesModelFilter(provider: string, modelId: string, modelFilter?: string): boolean { + if (!modelFilter) { + return true; + } + return modelId === modelFilter || `${provider}/${modelId}` === modelFilter; +} + +function matchesNameFilter(modelId: string, modelName: string, nameFilter?: string): boolean { + if (!nameFilter) { + return true; + } + const normalized = nameFilter.toLowerCase(); + return ( + modelId.toLowerCase() === normalized || + modelId.split('/').pop()?.toLowerCase() === normalized || + modelName.toLowerCase() === normalized + ); +} + +export const modelsSubcommand = createCommand({ + name: 'models', + aliases: ['list', 'ls'], + description: 'List AI Gateway models', + tags: ['read-only', 'fast'], + idempotent: true, + examples: [ + { command: getCommand('cloud aigateway models'), description: 'List all models' }, + { + command: getCommand('cloud aigateway models --provider openai'), + description: 'List OpenAI models', + }, + { + command: getCommand('cloud aigateway models --model anthropic/claude-opus-4-7'), + description: 'Show one model by id', + }, + ], + schema: { + options: z.object({ + model: z.string().optional().describe('show one model by full provider/id'), + provider: z.string().optional().describe('filter by provider'), + name: z + .string() + .optional() + .describe('show one model by id or display name with --provider'), + reasoning: z.boolean().optional().describe('only show reasoning models'), + input: z.string().optional().describe('filter by input modality, such as text or image'), + output: z.string().optional().describe('filter by output modality, such as text or image'), + ids: z.boolean().optional().describe('only print model ids'), + simple: z.boolean().optional().describe('print a compact model list'), + recommended: z.boolean().optional().describe('show recommended models for common uses'), + refreshModels: z + .boolean() + .optional() + .describe('refresh the cached AI Gateway model catalog'), + }), + response: ModelsResponseSchema, + }, + async handler(ctx) { + const service = createPublicAIGatewayService(ctx); + const profile = ctx.config?.name ?? 'default'; + const cacheKey = getAIGatewayUrl(ctx.region, ctx.config?.overrides); + const cached = ctx.opts.refreshModels + ? null + : await getCachedAIGatewayModels(profile, cacheKey); + const catalog = cached ?? (await service.listModels()); + if (!cached) { + await setCachedAIGatewayModels(profile, cacheKey, catalog); + } + const rows = Object.entries(catalog).flatMap(([provider, models]) => + models + .filter((model) => matchesProviderFilter(provider, model.id, ctx.opts.provider)) + .filter((model) => matchesModelFilter(provider, model.id, ctx.opts.model)) + .filter((model) => matchesNameFilter(model.id, model.name, ctx.opts.name)) + .filter((model) => !ctx.opts.reasoning || model.reasoning) + .filter((model) => !ctx.opts.input || model.input_modalities?.includes(ctx.opts.input)) + .filter( + (model) => !ctx.opts.output || model.output_modalities?.includes(ctx.opts.output) + ) + .map((model) => ({ + provider, + id: model.id, + name: model.name, + api: model.api, + reasoning: model.reasoning, + contextWindow: model.context_window, + maxOutputTokens: model.max_output_tokens, + })) + ); + const singleLookup = Boolean(ctx.opts.model || ctx.opts.name); + const selectedModel = singleLookup ? (rows[0] ?? null) : undefined; + + const agentOutput = isAgentOutputMode(); + if (ctx.options.json || agentOutput) { + if (agentOutput && !ctx.options.json) { + if (ctx.opts.ids) { + console.log( + JSON.stringify({ ids: rows.map((row) => row.id), count: rows.length }, null, 2) + ); + } else if (ctx.opts.recommended) { + console.log(JSON.stringify({ recommendations: getRecommendations(rows) }, null, 2)); + } else if (singleLookup) { + console.log( + JSON.stringify( + { model: selectedModel, models: rows, count: rows.length }, + null, + 2 + ) + ); + } else { + console.log(JSON.stringify({ models: rows, count: rows.length }, null, 2)); + } + } + } else { + if (rows.length === 0) { + tui.info('No AI Gateway models found'); + } else if (ctx.opts.ids) { + for (const row of rows) { + console.log(row.id); + } + } else if (ctx.opts.recommended) { + const recommendations = getRecommendations(rows).map((row) => ({ + Use: row.use, + Model: row.model, + Name: row.name, + })); + if (recommendations.length === 0) { + tui.info('No recommended AI Gateway models found'); + } else { + tui.table(recommendations, ['Use', 'Model', 'Name']); + } + } else if (ctx.opts.simple) { + tui.table( + rows.map((row) => ({ + Model: row.id, + Name: row.name, + })), + ['Model', 'Name'] + ); + } else { + tui.info(`Found ${rows.length} AI Gateway model(s):`); + tui.table( + rows.map((row) => ({ + Provider: row.provider, + Model: row.id, + Name: row.name, + API: row.api ?? '-', + Reasoning: row.reasoning ? 'yes' : 'no', + Context: row.contextWindow ?? '-', + })), + ['Provider', 'Model', 'Name', 'API', 'Reasoning', 'Context'] + ); + } + } + + return { models: rows, count: rows.length, model: selectedModel }; + }, +}); diff --git a/packages/cli/src/cmd/cloud/aigateway/util.ts b/packages/cli/src/cmd/cloud/aigateway/util.ts new file mode 100644 index 000000000..9a27432bb --- /dev/null +++ b/packages/cli/src/cmd/cloud/aigateway/util.ts @@ -0,0 +1,86 @@ +import { AIGatewayService, type Logger } from '@agentuity/core'; +import { createServerFetchAdapter, getServiceUrls } from '@agentuity/server'; +import * as tui from '../../../tui'; +import type { AuthData, Config, GlobalOptions, ProjectConfig } from '../../../types'; + +const defaultAIGatewayRegion = 'usc'; + +export function getAIGatewayUrl( + region?: string, + overrides?: { aigateway_url?: string } | null +): string { + if (process.env.AGENTUITY_AIGATEWAY_URL) { + return process.env.AGENTUITY_AIGATEWAY_URL; + } + if (overrides?.aigateway_url) { + return overrides.aigateway_url; + } + return getServiceUrls(region || process.env.AGENTUITY_REGION || defaultAIGatewayRegion) + .aigateway; +} + +export function createAIGatewayService(ctx: { + logger: Logger; + auth: AuthData; + region?: string; + project?: ProjectConfig; + config: Config | null; + options: GlobalOptions; +}) { + const orgId = + ctx.project?.orgId ?? + ctx.options.orgId ?? + (process.env.AGENTUITY_CLOUD_ORG_ID || ctx.config?.preferences?.orgId); + if (!orgId) { + tui.fatal( + 'Organization ID is required. Either run from a project directory or use --org-id flag.' + ); + } + + const adapter = createServerFetchAdapter( + { + headers: { + Authorization: `Bearer ${ctx.auth.apiKey}`, + 'x-agentuity-orgid': orgId, + }, + }, + ctx.logger + ); + + return new AIGatewayService(getAIGatewayUrl(ctx.region, ctx.config?.overrides), adapter); +} + +export function createPublicAIGatewayService(ctx: { + logger: Logger; + region?: string; + config: Config | null; +}) { + const adapter = createServerFetchAdapter({ headers: {} }, ctx.logger); + return new AIGatewayService(getAIGatewayUrl(ctx.region, ctx.config?.overrides), adapter); +} + +export function getCompletionText(response: unknown): string { + const choices = (response as { choices?: unknown }).choices; + const first = + Array.isArray(choices) && choices.length > 0 + ? (choices[0] as { message?: { content?: unknown }; text?: unknown; delta?: unknown }) + : undefined; + const content = + first?.message?.content ?? first?.text ?? (response as { content?: unknown }).content; + if (typeof content === 'string') { + return content; + } + if (Array.isArray(content)) { + return content + .map((part) => { + if (typeof part === 'string') return part; + if (part && typeof part === 'object' && 'text' in part) { + const text = (part as { text?: unknown }).text; + return typeof text === 'string' ? text : ''; + } + return ''; + }) + .join(''); + } + return ''; +} diff --git a/packages/cli/src/cmd/cloud/index.ts b/packages/cli/src/cmd/cloud/index.ts index 36c7479a6..d2e4d4c0b 100644 --- a/packages/cli/src/cmd/cloud/index.ts +++ b/packages/cli/src/cmd/cloud/index.ts @@ -14,6 +14,7 @@ import webhookCommand from './webhook'; import { agentCommand } from './agent'; import envCommand from './env'; import apikeyCommand from './apikey'; +import { aigatewayCommand } from './aigateway'; import oidcCommand from './oidc'; import streamCommand from './stream'; import vectorCommand from './vector'; @@ -41,6 +42,7 @@ export const command = createCommand({ ], subcommands: [ apikeyCommand, + aigatewayCommand, oidcCommand, keyvalueCommand, queueCommand, diff --git a/packages/cli/src/config.ts b/packages/cli/src/config.ts index ab506acca..948545438 100644 --- a/packages/cli/src/config.ts +++ b/packages/cli/src/config.ts @@ -226,6 +226,9 @@ export async function loadConfig( if (process.env.AGENTUITY_VECTOR_URL) { overrides.vector_url = process.env.AGENTUITY_VECTOR_URL; } + if (process.env.AGENTUITY_AIGATEWAY_URL) { + overrides.aigateway_url = process.env.AGENTUITY_AIGATEWAY_URL; + } if (process.env.AGENTUITY_STREAM_URL) { overrides.stream_url = process.env.AGENTUITY_STREAM_URL; } diff --git a/packages/cli/src/types.ts b/packages/cli/src/types.ts index 8d40915f1..6ee99cecf 100644 --- a/packages/cli/src/types.ts +++ b/packages/cli/src/types.ts @@ -41,6 +41,7 @@ export const ConfigSchema = zod.object({ kv_url: zod.url().optional().describe('Override keyvalue URL'), sandbox_url: zod.url().optional().describe('Override sandbox URL'), vector_url: zod.url().optional().describe('Override vector store URL'), + aigateway_url: zod.url().optional().describe('Override AI Gateway URL'), catalyst_url: zod.url().optional().describe('Override catalyst URL'), ion_url: zod.url().optional().describe('Override ion URL'), gravity_url: zod.url().optional().describe('Override gravity URL'), diff --git a/packages/cli/test/cmd/cloud/aigateway.test.ts b/packages/cli/test/cmd/cloud/aigateway.test.ts new file mode 100644 index 000000000..87b2ef06c --- /dev/null +++ b/packages/cli/test/cmd/cloud/aigateway.test.ts @@ -0,0 +1,560 @@ +import { mkdtemp, rm } from 'node:fs/promises'; +import { join } from 'node:path'; +import { tmpdir } from 'node:os'; +import { afterEach, describe, expect, test } from 'bun:test'; +import { createMinimalLogger } from '@agentuity/core'; +import { aigatewayCommand } from '../../../src/cmd/cloud/aigateway'; +import { combinePromptInput, completeSubcommand } from '../../../src/cmd/cloud/aigateway/complete'; +import { modelsSubcommand } from '../../../src/cmd/cloud/aigateway/models'; +import { getCompletionText } from '../../../src/cmd/cloud/aigateway/util'; + +let server: ReturnType | undefined; + +afterEach(() => { + server?.stop(true); + server = undefined; + delete process.env.AGENTUITY_AIGATEWAY_URL; + delete process.env.AGENTUITY_AIGATEWAY_MODEL; +}); + +function baseCtx(url: string) { + delete process.env.AGENTUITY_AIGATEWAY_URL; + return { + auth: { apiKey: 'sdk_test' }, + logger: createMinimalLogger(), + region: 'usc', + project: { orgId: 'org_test' }, + config: { overrides: { aigateway_url: url }, preferences: {} }, + options: { json: true }, + }; +} + +function completionModelCatalog(api = 'openai-responses') { + return Response.json({ + success: true, + data: { + openai: [{ id: 'gpt-4.1-mini', name: 'GPT 4.1 Mini', api }], + }, + }); +} + +describe('cloud aigateway command', () => { + test('registers expected subcommands', () => { + expect(aigatewayCommand.name).toBe('aigateway'); + expect(aigatewayCommand.aliases).toContain('ai-gateway'); + expect(aigatewayCommand.subcommands?.map((cmd) => cmd.name)).toEqual(['models', 'complete']); + expect(aigatewayCommand.requires?.auth).toBeUndefined(); + }); + + test('models subcommand is public', () => { + expect(modelsSubcommand.requires?.auth).toBeUndefined(); + expect(modelsSubcommand.requires?.region).toBeUndefined(); + expect(modelsSubcommand.idempotent).toBe(true); + expect(modelsSubcommand.schema?.response).toBeDefined(); + }); + + test('complete subcommand exposes prompt, model, stream, and convenience schemas', () => { + const shape = completeSubcommand.schema?.options?.def.shape; + expect(completeSubcommand.requires?.auth).toBe(true); + expect(completeSubcommand.requires?.region).toBeUndefined(); + expect(completeSubcommand.optional?.region).toBe(true); + expect(completeSubcommand.tags).toContain('uses-stdin'); + expect(shape?.model).toBeDefined(); + expect(shape?.prompt).toBeDefined(); + expect(shape?.stream).toBeDefined(); + expect(shape?.maxTokens).toBeDefined(); + expect(shape?.file).toBeDefined(); + expect(shape?.systemFile).toBeDefined(); + expect(shape?.save).toBeDefined(); + expect(shape?.format).toBeDefined(); + expect(shape?.stdinMode).toBeDefined(); + expect(shape?.cost).toBeDefined(); + }); + + test('extracts assistant text from OpenAI-compatible completion responses', () => { + expect( + getCompletionText({ + choices: [{ message: { role: 'assistant', content: 'hello' } }], + }) + ).toBe('hello'); + expect(getCompletionText({ choices: [{ text: 'fallback' }] })).toBe('fallback'); + expect( + getCompletionText({ + content: [{ type: 'text', text: 'anthropic text' }], + }) + ).toBe('anthropic text'); + }); + + test('combines explicit prompt and piped stdin by default', () => { + expect( + combinePromptInput({ + explicitPrompt: 'Summarize these records.', + stdinPrompt: '[{"name":"Ada"}]', + }) + ).toBe('Summarize these records.\n\n[{"name":"Ada"}]'); + expect( + combinePromptInput({ + explicitPrompt: 'Ignore this', + stdinPrompt: '[{"name":"Ada"}]', + stdinMode: 'replace', + }) + ).toBe('[{"name":"Ada"}]'); + }); + + test('models handler calls the configured gateway and returns flattened rows', async () => { + const requests: Request[] = []; + server = Bun.serve({ + port: 0, + fetch(request) { + requests.push(request); + return Response.json({ + success: true, + data: { + openai: [ + { + id: 'gpt-4.1-mini', + name: 'GPT 4.1 Mini', + api: 'openai-responses', + reasoning: false, + input_modalities: ['text'], + output_modalities: ['text'], + }, + { + id: 'gpt-4.1-vision', + name: 'GPT 4.1 Vision', + api: 'openai-responses', + reasoning: false, + input_modalities: ['text', 'image'], + output_modalities: ['text'], + }, + ], + }, + }); + }, + }); + + const result = await modelsSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { input: 'image' }, + args: {}, + } as never); + + expect(requests).toHaveLength(1); + expect(requests[0]!.method).toBe('GET'); + expect(requests[0]!.headers.get('authorization')).toBeNull(); + expect(requests[0]!.headers.get('x-agentuity-orgid')).toBeNull(); + expect(result.count).toBe(1); + expect(result.models[0]?.id).toBe('gpt-4.1-vision'); + }); + + test('models handler does not require auth, org, project, or region', async () => { + const requests: Request[] = []; + server = Bun.serve({ + port: 0, + fetch(request) { + requests.push(request); + return Response.json({ + success: true, + data: { + openai: [{ id: 'openai/gpt-4.1-mini', name: 'GPT 4.1 Mini' }], + }, + }); + }, + }); + + const result = await modelsSubcommand.handler({ + logger: createMinimalLogger(), + config: { + overrides: { aigateway_url: `http://127.0.0.1:${server.port}` }, + preferences: {}, + }, + options: { json: true }, + opts: {}, + args: {}, + } as never); + + expect(requests).toHaveLength(1); + expect(requests[0]!.headers.get('authorization')).toBeNull(); + expect(requests[0]!.headers.get('x-agentuity-orgid')).toBeNull(); + expect(result.count).toBe(1); + }); + + test('models handler filters by provider', async () => { + server = Bun.serve({ + port: 0, + fetch() { + return Response.json({ + success: true, + data: { + openai: [ + { + id: 'openai/gpt-4.1-mini', + name: 'GPT 4.1 Mini', + api: 'openai-responses', + }, + ], + anthropic: [ + { + id: 'anthropic/claude-sonnet-4-5-20250929', + name: 'Claude Sonnet 4.5', + api: 'anthropic-messages', + }, + ], + }, + }); + }, + }); + + const result = await modelsSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { provider: 'anthropic' }, + args: {}, + } as never); + + expect(result.count).toBe(1); + expect(result.models[0]?.provider).toBe('anthropic'); + expect(result.models[0]?.id).toBe('anthropic/claude-sonnet-4-5-20250929'); + }); + + test('models handler returns a single model by full model id', async () => { + server = Bun.serve({ + port: 0, + fetch() { + return Response.json({ + success: true, + data: { + anthropic: [ + { + id: 'claude-opus-4-7', + name: 'Claude Opus 4.7', + api: 'anthropic-messages', + }, + { + id: 'claude-sonnet-4-5-20250929', + name: 'Claude Sonnet 4.5', + api: 'anthropic-messages', + }, + ], + }, + }); + }, + }); + + const result = await modelsSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { model: 'anthropic/claude-opus-4-7' }, + args: {}, + } as never); + + expect(result.count).toBe(1); + expect(result.model?.provider).toBe('anthropic'); + expect(result.model?.id).toBe('claude-opus-4-7'); + }); + + test('models handler returns a single model by provider and name', async () => { + server = Bun.serve({ + port: 0, + fetch() { + return Response.json({ + success: true, + data: { + openai: [{ id: 'openai/gpt-4.1-mini', name: 'GPT 4.1 Mini' }], + anthropic: [{ id: 'anthropic/claude-opus-4-7', name: 'Claude Opus 4.7' }], + }, + }); + }, + }); + + const result = await modelsSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { provider: 'anthropic', name: 'Claude Opus 4.7' }, + args: {}, + } as never); + + expect(result.count).toBe(1); + expect(result.model?.id).toBe('anthropic/claude-opus-4-7'); + }); + + test('models subcommand exposes compact list options', () => { + const shape = modelsSubcommand.schema?.options?.def.shape; + expect(shape?.model).toBeDefined(); + expect(shape?.name).toBeDefined(); + expect(shape?.ids).toBeDefined(); + expect(shape?.simple).toBeDefined(); + expect(shape?.recommended).toBeDefined(); + }); + + test('complete handler posts an OpenAI-compatible chat completion request', async () => { + let body: unknown; + server = Bun.serve({ + port: 0, + async fetch(request) { + if (request.method === 'GET') { + return completionModelCatalog(); + } + body = await request.json(); + return Response.json( + { + id: 'chatcmpl_test', + model: 'gpt-4.1-mini', + choices: [{ message: { role: 'assistant', content: 'done' } }], + }, + { + headers: { + 'x-gateway-cost': '0.000456', + 'x-gateway-prompt-tokens': '12', + 'x-gateway-completion-tokens': '6', + }, + } + ); + }, + }); + + const result = await completeSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { model: 'gpt-4.1-mini', temperature: 0.2, maxTokens: 128, refreshModels: true }, + args: { prompt: 'Say done' }, + } as never); + + expect(body).toEqual({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say done' }], + temperature: 0.2, + max_tokens: 128, + }); + expect(result.text).toBe('done'); + expect(result.cost).toEqual({ + total: 0.000456, + promptTokens: 12, + completionTokens: 6, + }); + }); + + test('complete handler accepts prompt from --prompt option', async () => { + let body: unknown; + server = Bun.serve({ + port: 0, + async fetch(request) { + if (request.method === 'GET') { + return completionModelCatalog(); + } + body = await request.json(); + return Response.json({ + choices: [{ message: { role: 'assistant', content: 'from option' } }], + }); + }, + }); + + const result = await completeSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { model: 'gpt-4.1-mini', prompt: 'Prompt from option', refreshModels: true }, + args: {}, + } as never); + + expect(body).toEqual({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Prompt from option' }], + }); + expect(result.text).toBe('from option'); + }); + + test('complete handler accepts prompt from --file option', async () => { + let body: unknown; + const dir = await mkdtemp(join(tmpdir(), 'agentuity-aigateway-')); + const promptFile = join(dir, 'prompt.txt'); + await Bun.write(promptFile, 'Prompt from file\n'); + server = Bun.serve({ + port: 0, + async fetch(request) { + if (request.method === 'GET') { + return completionModelCatalog(); + } + body = await request.json(); + return Response.json({ + choices: [{ message: { role: 'assistant', content: 'from file' } }], + }); + }, + }); + + try { + const result = await completeSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { model: 'gpt-4.1-mini', file: promptFile, refreshModels: true }, + args: {}, + } as never); + + expect(body).toEqual({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Prompt from file' }], + }); + expect(result.text).toBe('from file'); + } finally { + await rm(dir, { recursive: true, force: true }); + } + }); + + test('complete handler uses messages payload for openai-compatible completions models', async () => { + let body: unknown; + server = Bun.serve({ + port: 0, + async fetch(request) { + if (request.method === 'GET') { + return Response.json({ + success: true, + data: { + poolside: [ + { + id: 'poolside/laguna-xs.2:free', + name: 'Laguna XS', + api: 'openai-completions', + }, + ], + }, + }); + } + body = await request.json(); + return Response.json({ + choices: [{ message: { role: 'assistant', content: 'compatible done' } }], + }); + }, + }); + + const result = await completeSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { + model: 'poolside/laguna-xs.2:free', + system: 'Be concise.', + refreshModels: true, + }, + args: { prompt: 'Say done' }, + } as never); + + expect(body).toEqual({ + model: 'poolside/laguna-xs.2:free', + messages: [ + { role: 'system', content: 'Be concise.' }, + { role: 'user', content: 'Say done' }, + ], + }); + expect(result.text).toBe('compatible done'); + }); + + test('complete handler uses AGENTUITY_AIGATEWAY_MODEL when model is omitted', async () => { + let body: unknown; + process.env.AGENTUITY_AIGATEWAY_MODEL = 'openai/gpt-env'; + server = Bun.serve({ + port: 0, + async fetch(request) { + if (request.method === 'GET') { + return completionModelCatalog(); + } + body = await request.json(); + return Response.json({ + choices: [{ message: { role: 'assistant', content: 'from env model' } }], + }); + }, + }); + + const result = await completeSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: {}, + args: { prompt: 'Say done' }, + } as never); + + expect(body).toEqual({ + model: 'openai/gpt-env', + messages: [{ role: 'user', content: 'Say done' }], + }); + expect(result.text).toBe('from env model'); + }); + + test('complete handler reads system prompt from --system-file and saves output', async () => { + let body: unknown; + const dir = await mkdtemp(join(tmpdir(), 'agentuity-aigateway-')); + const systemFile = join(dir, 'system.txt'); + const outputFile = join(dir, 'output.txt'); + await Bun.write(systemFile, 'Be concise.\n'); + server = Bun.serve({ + port: 0, + async fetch(request) { + if (request.method === 'GET') { + return completionModelCatalog(); + } + body = await request.json(); + return Response.json({ + choices: [{ message: { role: 'assistant', content: 'saved output' } }], + }); + }, + }); + + try { + const result = await completeSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { model: 'gpt-4.1-mini', systemFile, save: outputFile, refreshModels: true }, + args: { prompt: 'Say done' }, + } as never); + + expect(body).toEqual({ + model: 'gpt-4.1-mini', + messages: [ + { role: 'system', content: 'Be concise.' }, + { role: 'user', content: 'Say done' }, + ], + }); + expect(result.text).toBe('saved output'); + expect(await Bun.file(outputFile).text()).toBe('saved output'); + } finally { + await rm(dir, { recursive: true, force: true }); + } + }); + + test('complete handler streams token output when --stream is set', async () => { + let body: unknown; + server = Bun.serve({ + port: 0, + async fetch(request) { + if (request.method === 'GET') { + return completionModelCatalog(); + } + body = await request.json(); + return new Response( + [ + 'data: {"choices":[{"delta":{"content":"hel"}}]}', + '', + 'data: {"choices":[{"delta":{"content":"lo"}}]}', + '', + 'data: [DONE]', + '', + ].join('\n'), + { + headers: { + 'content-type': 'text/event-stream', + 'x-gateway-cost': '0.000789', + 'x-gateway-prompt-tokens': '20', + 'x-gateway-completion-tokens': '10', + }, + } + ); + }, + }); + + const result = await completeSubcommand.handler({ + ...baseCtx(`http://127.0.0.1:${server.port}`), + opts: { model: 'gpt-4.1-mini', stream: true, refreshModels: true }, + args: { prompt: 'Say hello' }, + } as never); + + expect(body).toEqual({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say hello' }], + stream: true, + }); + expect(result.text).toBe('hello'); + expect(result.response).toEqual({ stream: true, model: 'gpt-4.1-mini' }); + expect(result.cost).toEqual({ + total: 0.000789, + promptTokens: 20, + completionTokens: 10, + }); + }); +}); diff --git a/packages/cli/test/config/profile-creation.test.ts b/packages/cli/test/config/profile-creation.test.ts index bd914e651..3e732716c 100644 --- a/packages/cli/test/config/profile-creation.test.ts +++ b/packages/cli/test/config/profile-creation.test.ts @@ -24,6 +24,7 @@ const ENV_VARS_TO_CLEAR = [ 'AGENTUITY_KEYVALUE_URL', 'AGENTUITY_SANDBOX_URL', 'AGENTUITY_VECTOR_URL', + 'AGENTUITY_AIGATEWAY_URL', 'AGENTUITY_STREAM_URL', 'AGENTUITY_REGION', 'AGENTUITY_CLI_API_KEY', diff --git a/packages/coder-tui/package.json b/packages/coder-tui/package.json index 76ba1979d..abdf8cf7b 100644 --- a/packages/coder-tui/package.json +++ b/packages/coder-tui/package.json @@ -25,6 +25,8 @@ "prepublishOnly": "bun run clean && bun run build" }, "dependencies": { + "@agentuity/core": "workspace:*", + "@agentuity/server": "workspace:*", "@mariozechner/pi-coding-agent": "^0.72.1", "@mariozechner/pi-tui": "^0.72.1", "@sinclair/typebox": "^0.34.49" diff --git a/packages/coder-tui/src/aigateway.ts b/packages/coder-tui/src/aigateway.ts index 11fd1d2d1..22a8f5fff 100644 --- a/packages/coder-tui/src/aigateway.ts +++ b/packages/coder-tui/src/aigateway.ts @@ -10,6 +10,13 @@ import { delimiter, join } from 'node:path'; import { existsSync } from 'node:fs'; import { execFileSync } from 'node:child_process'; +import { createMinimalLogger, StructuredError } from '@agentuity/core'; +import { + AIGatewayService, + type AIGatewayModel, + type AIGatewayModels, +} from '@agentuity/core/aigateway'; +import { createServerFetchAdapter } from '@agentuity/server'; import type { ExtensionAPI, ProviderModelConfig } from '@mariozechner/pi-coding-agent'; export type KnownApi = @@ -24,8 +31,6 @@ export type KnownApi = | 'google-gemini-cli' | 'google-vertex'; -const MODEL_CATALOG_TIMEOUT_MS = 5_000; - const KNOWN_APIS = new Set([ 'openai-completions', 'mistral-conversations', @@ -39,33 +44,9 @@ const KNOWN_APIS = new Set([ 'google-vertex', ] satisfies KnownApi[]); -interface AIGatewayModels { - [key: string]: AIGatewayModel[]; -} - -interface AIGatewayModelResponse { - success: boolean; - data: AIGatewayModels; - message?: string; - error?: string; -} - -interface AIGatewayModel { - id: string; - name: string; - api: KnownApi; - reasoning: boolean; - input_modalities?: ('text' | 'image')[]; - context_window?: number; - max_output_tokens?: number; - pricing?: { - input: number; - output: number; - cached_input: number; - unit: 'per_million_tokens'; - currency: 'USD'; - }; -} +const AIGatewayModelFetchError = StructuredError('AIGatewayModelFetchError')<{ + cause?: unknown; +}>(); function getEnv(...keys: string[]): string | undefined { for (const key of keys) { @@ -132,8 +113,11 @@ async function fetchModels(): Promise { } } break; - } catch (_ex) { - // + } catch (error) { + throw new AIGatewayModelFetchError({ + message: 'Failed to fetch models from AI Gateway', + cause: error, + }); } } } @@ -155,46 +139,33 @@ async function fetchModels(): Promise { process.env.AGENTUITY_AIGATEWAY_ORGID = orgId; } - const controller = new AbortController(); - const timeout = setTimeout(() => controller.abort(), MODEL_CATALOG_TIMEOUT_MS); - try { - const response = await fetch(`${baseUrl}/models`, { signal: controller.signal }); - - if (!response.ok) { - console.warn( - `Failed to fetch models from AI Gateway: ${response.status} ${response.statusText}` - ); - return {}; - } - - const payload = (await response.json()) as AIGatewayModelResponse; - - if (!payload.success) { - console.warn(`Failed to load models. ${payload.message} ${payload}`); - } - - return payload.data; + const service = new AIGatewayService( + baseUrl, + createServerFetchAdapter({ headers: {} }, createMinimalLogger()) + ); + return await service.listModels(); } catch (error) { - if (error instanceof Error && error.name === 'AbortError') { - console.warn( - `Timed out fetching models from AI Gateway after ${MODEL_CATALOG_TIMEOUT_MS}ms` - ); - return {}; - } - console.warn('Failed to fetch models from AI Gateway:', error); - return {}; - } finally { - clearTimeout(timeout); + throw new AIGatewayModelFetchError({ + message: 'Failed to fetch models from AI Gateway', + cause: error, + }); } } +function sanitizeModalities(modalities: string[] | undefined): ('text' | 'image')[] { + const sanitized = (modalities ?? []).filter( + (modality): modality is 'text' | 'image' => modality === 'text' || modality === 'image' + ); + return sanitized.length > 0 ? sanitized : ['text']; +} + function toPiModel(m: AIGatewayModel): ProviderModelConfig { return { id: m.id, name: m.name, - reasoning: m.reasoning, - input: m.input_modalities as ('text' | 'image')[], + reasoning: m.reasoning ?? false, + input: sanitizeModalities(m.input_modalities), contextWindow: m.context_window ?? 40000, maxTokens: m.max_output_tokens ?? 64000, cost: { diff --git a/packages/coder-tui/tsconfig.json b/packages/coder-tui/tsconfig.json index dc1ef5181..5156d1abb 100644 --- a/packages/coder-tui/tsconfig.json +++ b/packages/coder-tui/tsconfig.json @@ -4,9 +4,14 @@ "composite": true, "outDir": "./dist", "rootDir": "./src", - "rewriteRelativeImportExtensions": true + "rewriteRelativeImportExtensions": true, + "paths": { + "@agentuity/core": ["../core/src"], + "@agentuity/core/*": ["../core/src/services/*"], + "@agentuity/server": ["../server/src"] + } }, "include": ["src/**/*"], "exclude": ["test/**/*"], - "references": [] + "references": [{ "path": "../core" }, { "path": "../server" }] } diff --git a/packages/core/src/env.d.ts b/packages/core/src/env.d.ts index 88cd902d5..477549b90 100644 --- a/packages/core/src/env.d.ts +++ b/packages/core/src/env.d.ts @@ -70,6 +70,12 @@ declare global { /** AI Gateway URL for model routing */ AGENTUITY_AIGATEWAY_URL?: string; + /** AI Gateway API key override */ + AGENTUITY_AIGATEWAY_KEY?: string; + + /** Default AI Gateway model for simple completions */ + AGENTUITY_AIGATEWAY_MODEL?: string; + /** Main API URL for Agentuity services */ AGENTUITY_API_URL?: string; diff --git a/packages/core/src/services/aigateway/api-reference.ts b/packages/core/src/services/aigateway/api-reference.ts new file mode 100644 index 000000000..90df52d2d --- /dev/null +++ b/packages/core/src/services/aigateway/api-reference.ts @@ -0,0 +1,167 @@ +import { z } from 'zod'; +import { + AIGatewayChatCompletionParamsSchema, + AIGatewayChatCompletionStreamParamsSchema, + AIGatewayChatCompletionSchema, + AIGatewayModelsResponseSchema, +} from './service.ts'; +import type { Service } from '../api-reference.ts'; + +const AIGatewayStreamCompletionSchema = z + .object({ + choices: z + .array( + z + .object({ + delta: z + .object({ + role: z + .string() + .optional() + .describe('Role for the streamed message delta.'), + content: z.string().optional().describe('Token or text delta.'), + }) + .optional() + .describe('Incremental assistant message content.'), + finish_reason: z + .string() + .nullable() + .optional() + .describe('Reason the model stopped generating, when available.'), + }) + .catchall(z.unknown()) + ) + .describe('Streamed completion choices.'), + }) + .catchall(z.unknown()) + .describe('A single Server-Sent Events data frame for streamed completions.'); + +const service: Service = { + name: 'AI Gateway', + slug: 'ai-gateway', + description: 'List supported LLM models and run routed AI Gateway completions', + host: 'aigateway', + hasPublicEndpoints: true, + endpoints: [ + { + id: 'list-models', + title: 'List Models', + method: 'GET', + path: '/models', + description: + 'List model metadata for LLM providers available through AI Gateway, grouped by provider.', + pathParams: [], + queryParams: [], + requestBody: null, + responseDescription: + 'JSON response with provider keys mapped to arrays of supported model metadata under the `data` envelope.', + responseFields: { schema: AIGatewayModelsResponseSchema }, + statuses: [ + { code: 200, description: 'Model catalog returned. Public — no auth required.' }, + ], + examplePath: '/models', + public: true, + }, + { + id: 'create-chat-completion', + title: 'Create Completion', + method: 'POST', + path: '/', + description: + 'Create a completion through the AI Gateway auto-router. The gateway routes by model and request shape, so chat `messages` and legacy `prompt` payloads are both supported.', + pathParams: [], + queryParams: [], + requestBody: { + description: + 'Completion request. Use `messages` for chat-compatible models and `prompt` for legacy OpenAI completions-compatible models. Additional provider-specific fields are passed through.', + fields: { schema: AIGatewayChatCompletionParamsSchema }, + }, + responseDescription: 'Provider-compatible completion response.', + responseHeaders: [ + { + name: 'X-Gateway-Cost', + description: + 'Estimated total gateway cost in USD, when billing metadata is available.', + }, + { + name: 'X-Gateway-Prompt-Tokens', + description: 'Prompt token count used for gateway billing.', + }, + { + name: 'X-Gateway-Completion-Tokens', + description: 'Completion token count used for gateway billing.', + }, + ], + responseFields: { schema: AIGatewayChatCompletionSchema, stripRequired: true }, + statuses: [ + { code: 200, description: 'Completion created' }, + { code: 400, description: 'Invalid completion request' }, + { code: 401, description: 'Unauthorized — invalid or missing API key' }, + { code: 402, description: 'Payment required — upgrade to a paid plan' }, + ], + examplePath: '/', + exampleBody: { + model: 'openai/gpt-4o-mini', + messages: [{ role: 'user', content: 'Say hello in one sentence.' }], + max_tokens: 64, + }, + }, + { + id: 'stream-chat-completion', + title: 'Stream Completion', + method: 'POST', + path: '/', + description: + 'Create a streaming completion through the AI Gateway auto-router. Set `stream: true` to receive Server-Sent Events token deltas.', + pathParams: [], + queryParams: [], + requestBody: { + description: 'Completion request with `stream` set to `true`.', + fields: { + schema: AIGatewayChatCompletionStreamParamsSchema, + overrides: { stream: { type: 'true' } }, + }, + }, + responseDescription: + 'Server-Sent Events stream. Each `data:` frame contains a provider-compatible delta payload. The stream ends with `data: [DONE]`.', + responseHeaders: [ + { + name: 'Trailer', + description: + 'Declares billing trailers such as `X-Gateway-Cost`, `X-Gateway-Prompt-Tokens`, and `X-Gateway-Completion-Tokens` for streamed responses.', + }, + { + name: 'X-Gateway-Cost', + description: + 'Estimated total gateway cost in USD. For streaming responses this may be delivered as an HTTP trailer after the body completes.', + }, + { + name: 'X-Gateway-Prompt-Tokens', + description: + 'Prompt token count used for gateway billing. For streaming responses this may be delivered as an HTTP trailer.', + }, + { + name: 'X-Gateway-Completion-Tokens', + description: + 'Completion token count used for gateway billing. For streaming responses this may be delivered as an HTTP trailer.', + }, + ], + responseFields: { schema: AIGatewayStreamCompletionSchema, stripRequired: true }, + statuses: [ + { code: 200, description: 'Streaming completion started' }, + { code: 400, description: 'Invalid completion request' }, + { code: 401, description: 'Unauthorized — invalid or missing API key' }, + { code: 402, description: 'Payment required — upgrade to a paid plan' }, + ], + examplePath: '/', + exampleHeaders: { Accept: 'text/event-stream' }, + exampleBody: { + model: 'openai/gpt-4o-mini', + messages: [{ role: 'user', content: 'Count to three.' }], + stream: true, + }, + }, + ], +}; + +export default service; diff --git a/packages/core/src/services/aigateway/index.ts b/packages/core/src/services/aigateway/index.ts new file mode 100644 index 000000000..465611af3 --- /dev/null +++ b/packages/core/src/services/aigateway/index.ts @@ -0,0 +1,24 @@ +export { + AIGatewayChatCompletionParamsSchema, + AIGatewayChatCompletionStreamParamsSchema, + AIGatewayChatCompletionSchema, + AIGatewayChatMessageSchema, + AIGatewayModelProviderSchema, + AIGatewayModelSchema, + AIGatewayModelsResponseSchema, + AIGatewayModelsSchema, + AIGatewayPricingSchema, + AIGatewayResponseMetadataSchema, + AIGatewayService, + type AIGatewayChatCompletion, + type AIGatewayChatCompletionParams, + type AIGatewayChatCompletionStreamParams, + type AIGatewayChatMessage, + type AIGatewayModel, + type AIGatewayModelProvider, + type AIGatewayModels, + type AIGatewayModelsResponse, + type AIGatewayPricing, + type AIGatewayResponseMetadata, + type AIGatewayStreamingCompletion, +} from './service.ts'; diff --git a/packages/core/src/services/aigateway/service.ts b/packages/core/src/services/aigateway/service.ts new file mode 100644 index 000000000..be8faf7a2 --- /dev/null +++ b/packages/core/src/services/aigateway/service.ts @@ -0,0 +1,355 @@ +import { z } from 'zod'; +import { StructuredError } from '../../error.ts'; +import { FetchAdapter } from '../adapter.ts'; +import { buildUrl, toServiceException, toPayload } from '../_util.ts'; + +const AIGatewayModelsResponseError = StructuredError('AIGatewayModelsResponseError')<{ + error?: string; + message?: string; +}>(); + +export const AIGatewayPricingSchema = z.object({ + input: z.number().describe('Input token price.'), + output: z.number().describe('Output token price.'), + cached_input: z.number().optional().describe('Cached input token price.'), + unit: z.string().describe('Pricing unit.'), + currency: z.string().describe('Pricing currency.'), +}); + +export type AIGatewayPricing = z.infer; + +export const AIGatewayModelProviderSchema = z.object({ + env: z.array(z.string()).optional().describe('Environment variables used by this provider.'), + api: z.string().optional().describe('Provider API URL.'), + doc: z.string().optional().describe('Provider documentation URL.'), + logo_url: z.string().optional().describe('Provider logo URL.'), +}); + +export type AIGatewayModelProvider = z.infer; + +export const AIGatewayModelSchema = z.object({ + id: z.string().describe('Model identifier.'), + name: z.string().describe('Display name.'), + created: z.number().optional().describe('Unix timestamp when the model was created.'), + api: z.string().optional().describe('Compatible provider API shape.'), + family: z.string().optional().describe('Model family.'), + context_window: z.number().optional().describe('Maximum context window.'), + max_output_tokens: z.number().optional().describe('Maximum output token count.'), + input_modalities: z.array(z.string()).optional().describe('Supported input modalities.'), + output_modalities: z.array(z.string()).optional().describe('Supported output modalities.'), + attachment: z.boolean().optional().describe('Whether the model supports attachments.'), + reasoning: z.boolean().optional().describe('Whether the model supports reasoning.'), + tool_call: z.boolean().optional().describe('Whether the model supports tool calls.'), + temperature: z.boolean().optional().describe('Whether the model supports temperature.'), + knowledge: z.string().optional().describe('Knowledge cutoff or label.'), + open_weights: z.boolean().optional().describe('Whether the model has open weights.'), + provider: AIGatewayModelProviderSchema.optional().describe('Provider metadata.'), + pricing: AIGatewayPricingSchema.optional().describe('Model pricing.'), +}); + +export type AIGatewayModel = z.infer; + +export const AIGatewayModelsSchema = z.record(z.string(), z.array(AIGatewayModelSchema)); +export type AIGatewayModels = z.infer; + +export const AIGatewayModelsResponseSchema = z.object({ + success: z.boolean(), + data: AIGatewayModelsSchema.optional(), + message: z.string().optional(), + error: z.string().optional(), +}); + +export type AIGatewayModelsResponse = z.infer; + +export const AIGatewayChatMessageSchema = z.object({ + role: z.enum(['system', 'developer', 'user', 'assistant', 'tool']), + content: z + .union([ + z.string(), + z.array( + z + .object({ + type: z.string(), + }) + .catchall(z.unknown()) + ), + z.null(), + ]) + .optional(), + name: z.string().optional(), + tool_call_id: z.string().optional(), + tool_calls: z.array(z.unknown()).optional(), +}); + +export type AIGatewayChatMessage = z.infer; + +const missingCompletionInputMessage = 'either prompt or messages must be provided'; + +function hasCompletionInput(params: { prompt?: string | string[]; messages?: unknown[] }): boolean { + if (params.messages && params.messages.length > 0) { + return true; + } + if (typeof params.prompt === 'string') { + return params.prompt.trim().length > 0; + } + if (Array.isArray(params.prompt)) { + return params.prompt.length > 0 && params.prompt.every((item) => item.trim().length > 0); + } + return false; +} + +export const AIGatewayChatCompletionParamsSchema = z + .object({ + model: z.string().describe('Model to use for the completion.'), + messages: z.array(AIGatewayChatMessageSchema).optional().describe('Messages to complete.'), + prompt: z + .union([z.string(), z.array(z.string())]) + .optional() + .describe('Prompt to complete.'), + temperature: z.number().optional(), + top_p: z.number().optional(), + max_tokens: z.number().optional(), + stream: z.boolean().optional(), + stop: z.union([z.string(), z.array(z.string())]).optional(), + }) + .catchall(z.unknown()) + .superRefine((params, ctx) => { + if (!hasCompletionInput(params)) { + ctx.addIssue({ + code: 'custom', + message: missingCompletionInputMessage, + path: ['messages'], + }); + } + }); + +export type AIGatewayChatCompletionParams = z.infer; + +export const AIGatewayChatCompletionStreamParamsSchema = + AIGatewayChatCompletionParamsSchema.safeExtend({ + stream: z.literal(true).describe('Enable Server-Sent Events streaming.'), + }); + +export type AIGatewayChatCompletionStreamParams = z.infer< + typeof AIGatewayChatCompletionStreamParamsSchema +>; + +export const AIGatewayChatCompletionSchema = z + .object({ + id: z.string().optional(), + object: z.string().optional(), + created: z.number().optional(), + model: z.string().optional(), + choices: z.array(z.unknown()).optional(), + usage: z.unknown().optional(), + agentuity: z + .object({ + headers: z + .record(z.string(), z.string()) + .optional() + .describe('AI Gateway response headers captured from the HTTP response.'), + cost: z + .object({ + total: z.number().optional().describe('Total estimated gateway cost in USD.'), + promptTokens: z + .number() + .optional() + .describe('Prompt token count used for gateway billing.'), + completionTokens: z + .number() + .optional() + .describe('Completion token count used for gateway billing.'), + }) + .optional() + .describe('Parsed AI Gateway cost information when available.'), + }) + .optional() + .describe('Agentuity AI Gateway metadata.'), + }) + .catchall(z.unknown()); + +export type AIGatewayChatCompletion = z.infer; + +export const AIGatewayResponseMetadataSchema = z.object({ + headers: z.record(z.string(), z.string()).optional(), + cost: z + .object({ + total: z.number().optional(), + promptTokens: z.number().optional(), + completionTokens: z.number().optional(), + }) + .optional(), +}); + +export type AIGatewayResponseMetadata = z.infer; + +export type AIGatewayStreamingCompletion = { + stream: ReadableStream; + metadata: Promise; +}; + +function parseNumber(value: string | undefined): number | undefined { + if (value === undefined || value.trim() === '') { + return undefined; + } + const parsed = Number(value); + return Number.isFinite(parsed) ? parsed : undefined; +} + +function extractGatewayMetadataFromHeaders(headers: Headers): AIGatewayResponseMetadata { + const captured: Record = {}; + for (const [key, value] of headers.entries()) { + const lower = key.toLowerCase(); + if ( + lower.startsWith('x-gateway-') || + (lower.startsWith('x-agentuity-') && + (lower.includes('cost') || lower.includes('token') || lower.includes('usage'))) + ) { + captured[lower] = value; + } + } + + const total = parseNumber(captured['x-gateway-cost']); + const promptTokens = parseNumber(captured['x-gateway-prompt-tokens']); + const completionTokens = parseNumber(captured['x-gateway-completion-tokens']); + const cost = + total !== undefined || promptTokens !== undefined || completionTokens !== undefined + ? { total, promptTokens, completionTokens } + : undefined; + + return { + ...(Object.keys(captured).length > 0 ? { headers: captured } : {}), + ...(cost ? { cost } : {}), + }; +} + +async function extractGatewayMetadata(response: Response): Promise { + const metadata = extractGatewayMetadataFromHeaders(response.headers); + const trailers = (response as Response & { trailers?: Promise }).trailers; + if (trailers) { + try { + const trailerMetadata = extractGatewayMetadataFromHeaders(await trailers); + const cost = + metadata.cost || trailerMetadata.cost + ? { ...(metadata.cost ?? {}), ...(trailerMetadata.cost ?? {}) } + : undefined; + return { + headers: { ...metadata.headers, ...trailerMetadata.headers }, + ...(cost ? { cost } : {}), + }; + } catch { + // Some runtimes expose a trailers promise but reject when trailers are unavailable. + } + } + return metadata; +} + +function attachGatewayMetadata>( + payload: T, + metadata: AIGatewayResponseMetadata +): T { + if (!metadata.headers && !metadata.cost) { + return payload; + } + return { + ...payload, + agentuity: { + ...(typeof payload.agentuity === 'object' && payload.agentuity !== null + ? payload.agentuity + : {}), + ...metadata, + }, + }; +} + +export class AIGatewayService { + constructor( + readonly baseUrl: string, + readonly adapter: FetchAdapter + ) {} + + async listModels(): Promise { + const method = 'GET'; + const url = buildUrl(this.baseUrl, '/models'); + const response = await this.adapter.invoke(url, { + method, + telemetry: { name: 'aigateway.models.list' }, + }); + if (!response.ok) { + throw await toServiceException(method, url, response.response); + } + const payload = AIGatewayModelsResponseSchema.parse(response.data); + if (!payload.success) { + throw new AIGatewayModelsResponseError({ + message: payload.error || payload.message || 'AI Gateway failed to list models', + error: payload.error, + }); + } + if (!payload.data) { + throw new AIGatewayModelsResponseError({ + message: 'AI Gateway model response did not include data', + }); + } + return payload.data; + } + + async complete(params: AIGatewayChatCompletionParams): Promise { + const method = 'POST'; + const url = buildUrl(this.baseUrl, '/'); + const [body, contentType] = await toPayload( + AIGatewayChatCompletionParamsSchema.parse(params) + ); + const response = await this.adapter.invoke(url, { + method, + body, + contentType, + telemetry: { name: 'aigateway.completions.create' }, + }); + if (!response.ok) { + throw await toServiceException(method, url, response.response); + } + const payload = attachGatewayMetadata( + response.data as Record, + await extractGatewayMetadata(response.response) + ); + return AIGatewayChatCompletionSchema.parse(payload); + } + + async streamComplete( + params: AIGatewayChatCompletionParams + ): Promise> { + return (await this.streamCompleteWithMetadata(params)).stream; + } + + async streamCompleteWithMetadata( + params: AIGatewayChatCompletionParams + ): Promise { + const method = 'POST'; + const url = buildUrl(this.baseUrl, '/'); + const [body, contentType] = await toPayload( + AIGatewayChatCompletionParamsSchema.parse({ ...params, stream: true }) + ); + const response = await this.adapter.invoke(url, { + method, + body, + contentType, + headers: { Accept: 'text/event-stream' }, + binary: true, + telemetry: { name: 'aigateway.completions.stream' }, + }); + if (!response.ok) { + throw await toServiceException(method, url, response.response); + } + if (!response.response.body) { + throw await toServiceException( + method, + url, + new Response('Streaming response did not include a body', { status: 502 }) + ); + } + return { + stream: response.response.body, + metadata: extractGatewayMetadata(response.response), + }; + } +} diff --git a/packages/core/src/services/api-reference.ts b/packages/core/src/services/api-reference.ts index 27061288f..a440b5201 100644 --- a/packages/core/src/services/api-reference.ts +++ b/packages/core/src/services/api-reference.ts @@ -60,6 +60,7 @@ interface Endpoint { exampleBody?: string | object; exampleHeaders?: Record; ttlNote?: string; + public?: boolean; } interface Service { diff --git a/packages/core/src/services/config.ts b/packages/core/src/services/config.ts index 6c0d2fa34..933549c35 100644 --- a/packages/core/src/services/config.ts +++ b/packages/core/src/services/config.ts @@ -6,6 +6,7 @@ export const ServiceUrlsSchema = z keyvalue: z.string().describe('URL for the key-value storage service.'), stream: z.string().describe('URL for the stream service.'), vector: z.string().describe('URL for the vector storage service.'), + aigateway: z.string().describe('URL for the AI Gateway service.'), catalyst: z.string().describe('URL for the Catalyst API gateway.'), otel: z.string().describe('URL for the OpenTelemetry collector.'), sandbox: z.string().describe('URL for the sandbox service.'), @@ -42,6 +43,7 @@ export function getServiceUrls(region?: string): ServiceUrls { keyvalue: getEnv('AGENTUITY_KEYVALUE_URL') || transportUrl, stream: getEnv('AGENTUITY_STREAM_URL') || buildRegionalURL(resolvedRegion, 'streams'), vector: getEnv('AGENTUITY_VECTOR_URL') || transportUrl, + aigateway: getEnv('AGENTUITY_AIGATEWAY_URL') || buildRegionalURL(resolvedRegion, 'aigateway'), catalyst: getEnv('AGENTUITY_CATALYST_URL') || transportUrl, otel: getEnv('AGENTUITY_OTLP_URL') || buildRegionalURL(resolvedRegion, 'otel'), sandbox: getEnv('AGENTUITY_SANDBOX_URL') || transportUrl, diff --git a/packages/core/src/services/index.ts b/packages/core/src/services/index.ts index 56aae5fc3..c841cc844 100644 --- a/packages/core/src/services/index.ts +++ b/packages/core/src/services/index.ts @@ -1,4 +1,5 @@ export * from './adapter.ts'; +export * from './aigateway/index.ts'; export * from './auth/index.ts'; export * from './email/index.ts'; export * from './exception.ts'; diff --git a/packages/core/test/aigateway.test.ts b/packages/core/test/aigateway.test.ts new file mode 100644 index 000000000..23722d15b --- /dev/null +++ b/packages/core/test/aigateway.test.ts @@ -0,0 +1,173 @@ +import { describe, expect, test } from 'bun:test'; +import { createMockAdapter } from '@agentuity/test-utils'; +import { + AIGatewayChatCompletionParamsSchema, + AIGatewayService, +} from '../src/services/aigateway/index.ts'; + +describe('AIGatewayService', () => { + const baseUrl = 'https://aigateway.example.com'; + + test('requires prompt or messages for completion params', () => { + expect(AIGatewayChatCompletionParamsSchema.safeParse({ model: 'gpt-4.1-mini' }).success).toBe( + false + ); + expect( + AIGatewayChatCompletionParamsSchema.safeParse({ + model: 'gpt-4.1-mini', + prompt: ' ', + }).success + ).toBe(false); + expect( + AIGatewayChatCompletionParamsSchema.safeParse({ + model: 'gpt-4.1-mini', + prompt: ['Say hello'], + }).success + ).toBe(true); + }); + + test('lists models from the gateway catalog', async () => { + const { adapter, calls } = createMockAdapter([ + { + ok: true, + data: { + success: true, + data: { + openai: [ + { + id: 'gpt-4.1-mini', + name: 'GPT 4.1 Mini', + reasoning: false, + input_modalities: ['text'], + output_modalities: ['text'], + provider: { api: 'https://api.openai.com' }, + pricing: { + input: 0.4, + output: 1.6, + unit: 'per_million_tokens', + currency: 'USD', + }, + }, + ], + }, + }, + }, + ]); + const service = new AIGatewayService(baseUrl, adapter); + + const models = await service.listModels(); + + expect(calls).toHaveLength(1); + expect(calls[0]?.url).toBe(`${baseUrl}/models`); + expect(calls[0]?.options.method).toBe('GET'); + expect(models.openai?.[0]?.id).toBe('gpt-4.1-mini'); + }); + + test('creates completions through the AI Gateway auto-router endpoint', async () => { + const { adapter, calls } = createMockAdapter([ + { + ok: true, + data: { + id: 'chatcmpl_123', + model: 'gpt-4.1-mini', + choices: [{ message: { role: 'assistant', content: 'Hello' } }], + }, + headers: { + 'x-gateway-cost': '0.000123', + 'x-gateway-prompt-tokens': '10', + 'x-gateway-completion-tokens': '5', + }, + }, + ]); + const service = new AIGatewayService(baseUrl, adapter); + + const completion = await service.complete({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say hello' }], + temperature: 0.2, + }); + + expect(calls).toHaveLength(1); + expect(calls[0]?.url).toBe(`${baseUrl}/`); + expect(calls[0]?.options.method).toBe('POST'); + expect(calls[0]?.options.contentType).toBe('application/json'); + expect(JSON.parse(String(calls[0]?.options.body))).toEqual({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say hello' }], + temperature: 0.2, + }); + expect(completion.id).toBe('chatcmpl_123'); + expect(completion.agentuity?.cost).toEqual({ + total: 0.000123, + promptTokens: 10, + completionTokens: 5, + }); + expect(completion.agentuity?.headers?.['x-gateway-cost']).toBe('0.000123'); + }); + + test('streams completions through the AI Gateway auto-router endpoint', async () => { + const { adapter, calls } = createMockAdapter([ + { + ok: true, + data: undefined, + headers: { 'content-type': 'text/event-stream' }, + body: 'data: {"choices":[{"delta":{"content":"Hello"}}]}\n\n', + }, + ]); + const service = new AIGatewayService(baseUrl, adapter); + + const stream = await service.streamComplete({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say hello' }], + }); + + expect(calls).toHaveLength(1); + expect(calls[0]?.url).toBe(`${baseUrl}/`); + expect(calls[0]?.options.method).toBe('POST'); + expect(calls[0]?.options.contentType).toBe('application/json'); + expect(calls[0]?.options.headers).toEqual({ Accept: 'text/event-stream' }); + expect(calls[0]?.options.binary).toBe(true); + expect(JSON.parse(String(calls[0]?.options.body))).toEqual({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say hello' }], + stream: true, + }); + expect(stream).toBeInstanceOf(ReadableStream); + }); + + test('streams chat completions with gateway metadata', async () => { + const { adapter } = createMockAdapter([ + { + ok: true, + data: undefined, + headers: { + 'content-type': 'text/event-stream', + 'x-gateway-cost': '0.000234', + 'x-gateway-prompt-tokens': '11', + 'x-gateway-completion-tokens': '7', + }, + body: 'data: {"choices":[{"delta":{"content":"Hello"}}]}\n\n', + }, + ]); + const service = new AIGatewayService(baseUrl, adapter); + + const completion = await service.streamCompleteWithMetadata({ + model: 'gpt-4.1-mini', + messages: [{ role: 'user', content: 'Say hello' }], + }); + + expect(completion.stream).toBeInstanceOf(ReadableStream); + expect(await completion.metadata).toEqual({ + headers: { + 'x-gateway-cost': '0.000234', + 'x-gateway-prompt-tokens': '11', + 'x-gateway-completion-tokens': '7', + }, + cost: { + total: 0.000234, + promptTokens: 11, + completionTokens: 7, + }, + }); + }); +}); diff --git a/packages/server/README.md b/packages/server/README.md index 54053fb5d..be5a2805a 100644 --- a/packages/server/README.md +++ b/packages/server/README.md @@ -25,6 +25,7 @@ const urls: ServiceUrls = getServiceUrls(region); console.log(urls.keyvalue); // https://agentuity.ai (or AGENTUITY_KEYVALUE_URL) console.log(urls.stream); // https://streams.agentuity.cloud (or AGENTUITY_STREAM_URL) console.log(urls.vector); // https://agentuity.ai (or AGENTUITY_VECTOR_URL) +console.log(urls.aigateway); // https://aigateway-usc.agentuity.cloud (or AGENTUITY_AIGATEWAY_URL) ``` ### Server Fetch Adapter diff --git a/packages/server/src/config.ts b/packages/server/src/config.ts index 596f11346..d3e00120f 100644 --- a/packages/server/src/config.ts +++ b/packages/server/src/config.ts @@ -2,6 +2,7 @@ export interface ServiceUrls { keyvalue: string; stream: string; vector: string; + aigateway: string; catalyst: string; otel: string; sandbox: string; @@ -35,6 +36,8 @@ export function getServiceUrls(region?: string): ServiceUrls { keyvalue: process.env.AGENTUITY_KEYVALUE_URL || transportUrl, stream: process.env.AGENTUITY_STREAM_URL || buildRegionalURL(resolvedRegion, 'streams'), vector: process.env.AGENTUITY_VECTOR_URL || transportUrl, + aigateway: + process.env.AGENTUITY_AIGATEWAY_URL || buildRegionalURL(resolvedRegion, 'aigateway'), catalyst: process.env.AGENTUITY_CATALYST_URL || transportUrl, otel: process.env.AGENTUITY_OTLP_URL || buildRegionalURL(resolvedRegion, 'otel'), sandbox: process.env.AGENTUITY_SANDBOX_URL || transportUrl, diff --git a/packages/server/test/config.test.ts b/packages/server/test/config.test.ts index cd1e21b9a..64a44ce0c 100644 --- a/packages/server/test/config.test.ts +++ b/packages/server/test/config.test.ts @@ -36,6 +36,7 @@ describe('getServiceUrls', () => { delete process.env.AGENTUITY_OBJECTSTORE_URL; delete process.env.AGENTUITY_STREAM_URL; delete process.env.AGENTUITY_VECTOR_URL; + delete process.env.AGENTUITY_AIGATEWAY_URL; delete process.env.AGENTUITY_CATALYST_URL; delete process.env.AGENTUITY_OTLP_URL; }); @@ -51,6 +52,7 @@ describe('getServiceUrls', () => { const urls = getServiceUrls(); expect(urls.catalyst).toBe('https://catalyst-us-west.agentuity.cloud'); expect(urls.stream).toBe('https://streams-us-west.agentuity.cloud'); + expect(urls.aigateway).toBe('https://aigateway-us-west.agentuity.cloud'); }); test('should build URLs for us-east region', () => { @@ -58,12 +60,14 @@ describe('getServiceUrls', () => { expect(urls.catalyst).toBe('https://catalyst-us-east.agentuity.cloud'); expect(urls.keyvalue).toBe('https://catalyst-us-east.agentuity.cloud'); expect(urls.stream).toBe('https://streams-us-east.agentuity.cloud'); + expect(urls.aigateway).toBe('https://aigateway-us-east.agentuity.cloud'); }); test('should use agentuity.io for local region', () => { const urls = getServiceUrls('local'); expect(urls.catalyst).toBe('https://catalyst.agentuity.io'); expect(urls.stream).toBe('https://streams.agentuity.io'); + expect(urls.aigateway).toBe('https://aigateway.agentuity.io'); }); test('should override with AGENTUITY_TRANSPORT_URL', () => { @@ -77,9 +81,11 @@ describe('getServiceUrls', () => { test('should override individual service URLs', () => { process.env.AGENTUITY_SANDBOX_URL = 'https://custom-sandbox.example.com'; process.env.AGENTUITY_KEYVALUE_URL = 'https://custom-kv.example.com'; + process.env.AGENTUITY_AIGATEWAY_URL = 'https://custom-ai.example.com'; const urls = getServiceUrls('us-east'); expect(urls.keyvalue).toBe('https://custom-kv.example.com'); + expect(urls.aigateway).toBe('https://custom-ai.example.com'); expect(urls.catalyst).toBe('https://catalyst-us-east.agentuity.cloud'); expect(urls.sandbox).toBe('https://custom-sandbox.example.com'); }); diff --git a/tsconfig.json b/tsconfig.json index 6e8a61812..abec9854f 100644 --- a/tsconfig.json +++ b/tsconfig.json @@ -2,6 +2,7 @@ "files": [], "references": [ { "path": "./packages/auth" }, + { "path": "./packages/aigateway" }, { "path": "./packages/claude-code" }, { "path": "./packages/cli" }, { "path": "./packages/coder" },