{
  "opencollection": "1.0.0",
  "info": {
    "name": "LangSmith access_policies gateway API",
    "version": "0.1.0"
  },
  "items": [
    {
      "info": {
        "name": "gateway",
        "type": "folder"
      },
      "items": [
        {
          "info": {
            "name": "Proxy Anthropic Messages through the LLM Gateway",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "/anthropic/v1/messages",
            "body": {
              "type": "json",
              "data": "{}"
            },
            "auth": {
              "type": "apikey",
              "key": "X-API-Key",
              "value": "{{X-API-Key}}",
              "placement": "header"
            }
          },
          "docs": "Forwards an Anthropic Messages API request to api.anthropic.com.\nThe gateway injects the workspace's stored ANTHROPIC_API_KEY\nunless the request carries a Claude-Code-style OAuth bearer\n(`Authorization: Bearer sk-ant-oat01-...`), in which case the\nbearer is forwarded as-is and the workspace key is not used.\nSpend-cap and guard policies apply in both modes.\n\nStreaming (`stream: true`) responses are forwarded as\n`text/event-stream` chunks. Non-streaming responses are\nforwarded as `application/json"
        },
        {
          "info": {
            "name": "Proxy Google Gemini generateContent through the LLM Gateway",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "/gemini/v1beta/models/:model",
            "params": [
              {
                "name": "model",
                "value": "",
                "type": "path",
                "description": "Model + action, e.g. `gemini-2.0-flash:generateContent` or `gemini-2.0-flash:streamGenerateContent`"
              }
            ],
            "body": {
              "type": "json",
              "data": "{}"
            },
            "auth": {
              "type": "apikey",
              "key": "X-API-Key",
              "value": "{{X-API-Key}}",
              "placement": "header"
            }
          },
          "docs": "Forwards a Gemini v1beta `generateContent` or\n`streamGenerateContent` request to\ngenerativelanguage.googleapis.com. The gateway injects the\nworkspace's stored GOOGLE_API_KEY as the `x-goog-api-key`\nheader; any client-supplied `?key=` query parameter is\nstripped before forwarding so the workspace credential is the\nonly one in play. Spend-cap and guard policies apply.\n\n`streamGenerateContent` responses are forwarded as\n`text/event-stream`; `generateContent` responses are\nforwarded as `application/"
        },
        {
          "info": {
            "name": "Proxy OpenAI Chat Completions through the LLM Gateway",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "/openai/v1/chat/completions",
            "body": {
              "type": "json",
              "data": "{}"
            },
            "auth": {
              "type": "apikey",
              "key": "X-API-Key",
              "value": "{{X-API-Key}}",
              "placement": "header"
            }
          },
          "docs": "Forwards an OpenAI Chat Completions request to api.openai.com.\nThe request body is the unmodified OpenAI request shape; the\ngateway injects the workspace's stored OPENAI_API_KEY, applies\nspend-cap and guard policies, and emits a LangSmith trace span\nfor the call.\n\nStreaming (`stream: true`) responses are forwarded as\n`text/event-stream` chunks. Non-streaming responses are\nforwarded as `application/json`."
        },
        {
          "info": {
            "name": "Proxy OpenAI Responses through the LLM Gateway",
            "type": "http"
          },
          "http": {
            "method": "POST",
            "url": "/openai/v1/responses",
            "body": {
              "type": "json",
              "data": "{}"
            },
            "auth": {
              "type": "apikey",
              "key": "X-API-Key",
              "value": "{{X-API-Key}}",
              "placement": "header"
            }
          },
          "docs": "Forwards an OpenAI Responses API request to api.openai.com.\nSame auth, policy, and tracing semantics as Chat Completions —\nthe workspace's OPENAI_API_KEY is injected by the gateway and\nthe request body is forwarded verbatim."
        }
      ]
    }
  ],
  "bundled": true
}