Summary

Describe the problem and fix in 2–5 bullets:
This is just a small extension to this PR: #49543
We now use the "azure-openai-responses" implementation for azure openai custom provider endpoint. A manually configured azure openai endpoint with api "azure-openai-responses" fails as the enum MODEL_APIS in types.models.ts does not have an entry for "azure-openai-responses"

Response sanitization, etc not needed for azure-openai-responses as we have done for openai-responses, verified it manually

• Problem:
• Why it matters: pi-ai has a dedicated handler for azure-openai endpoints named "azure-openai-responses" to take care of azure-openai responses API calls, we might as well use it over "openai-responses" for azure openai endpoints as it has some specific handling over "openai-responses" which users might hit for some edge cases.
• What changed: For custom azure openai endpoints, we were earlier using "openai-responses" we now use "azure-openai-responses"
• What did NOT change (scope boundary): Scope is only the config generated for azure openai endpoints through custom provider onboarding.

Change Type (select all)

• Bug fix
• Feature
• Refactor
• Docs
• Security hardening
• Chore/infra

Scope (select all touched areas)

• Gateway / orchestration
• Skills / tool execution
• Auth / tokens
• Memory / storage
• Integrations
• API / contracts
• UI / DX
• CI/CD / infra

Linked Issue/PR

• Closes #
• Related: fix(provider): tui/agent etc not working when azure openai endpoint is onboarded through custom provider #49543 & [Bug]: tui/agent etc not working when azure openai endpoint is onboarded #48939

User-visible / Behavior Changes

List user-visible changes (including defaults/config).
The generated config now has api "azure-openai-responses" rather than "openai-responses" for custom provider azure openai endpoints.

Security Impact (required)

• New permissions/capabilities? No
• Secrets/tokens handling changed? No
• New/changed network calls? No
• Command/tool execution surface changed? No
• Data access scope changed? No
• If any Yes, explain risk + mitigation:

Repro + Verification

Manually verified that after the change, tui and agent with tool calls work with a configured azure openai endpoint

Manually verified that after the change, the api field in focus generates "azure-openai-responses"

Environment

• OS: Ubuntu 25.10/Ubuntu 24.04
• Runtime/container: Tested in both Azure VM & Docker container
• Model/provider: Azure OpenAI gpt-5.2-chat & gpt-5.3-chat
• Integration/channel (if any):
• Relevant config (redacted):

{
  "wizard": {
    "lastRunAt": "2026-03-20T04:33:02.576Z",
    "lastRunVersion": "2026.3.14",
    "lastRunCommand": "onboard",
    "lastRunMode": "local"
  },
  "models": {
    "mode": "merge",
    "providers": {
      "custom-kkarmakar-ai-eus2-openai-azure-com": {
        "baseUrl": "https://resource.openai.azure.com/openai/v1",
        "apiKey": "key1234",
        "api": "azure-openai-responses",
        "headers": {
          "api-key": "key1234"
        },
        "authHeader": false,
        "models": [
          {
            "id": "gpt-5.3-chat",
            "name": "gpt-5.3-chat (Custom Provider)",
            "reasoning": true,
            "input": [
              "text",
              "image"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 400000,
            "maxTokens": 16384,
            "compat": {
              "supportsStore": false
            }
          }
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "custom-resource-openai-azure-com/gpt-5.3-chat"
      },
      "models": {
        "custom-resource-openai-azure-com/gpt-5.3-chat": {
          "params": {
            "thinking": "medium"
          }
        }
      },
      "workspace": "/root/.openclaw/workspace"
    }
  },
  "tools": {
    "profile": "coding"
  },
  "commands": {
    "native": "auto",
    "nativeSkills": "auto",
    "restart": true,
    "ownerDisplay": "raw"
  },
  "session": {
    "dmScope": "per-channel-peer"
  },
  "gateway": {
    "port": 18789,
    "mode": "local",
    "bind": "loopback",
    "auth": {
      "mode": "token",
      "token": "0e8dbc50a15d49e09a70c4594b0f1a0abcbce8f6e1ebc51d"
    },
    "tailscale": {
      "mode": "off",
      "resetOnExit": false
    },
    "nodes": {
      "denyCommands": [
        "camera.snap",
        "camera.clip",
        "screen.record",
        "contacts.add",
        "calendar.add",
        "reminders.add",
        "sms.send"
      ]
    }
  },
  "meta": {
    "lastTouchedVersion": "2026.3.14",
    "lastTouchedAt": "2026-03-20T04:33:02.586Z"
  }
}

Steps

1. Install openclaw
2. Onboard an azure openai endpoint using custom provider option or onboard an azure openai endpoint with api "azure-openai-responses" manually

Expected

TUI/Agent should work

Actual

Currently, without the change, onboarding wizard generates a config with "openai-responses" api fo azure openai responses which works but we dont know if any use case will break as pi-ai has a dedicated "azure-openai-responses" specifically for azure openai endpoints.
A manually configured azure openai endpoint with api "azure-openai-responses" fails as the enum MODEL_APIS in types.models.ts does not have an entry for "azure-openai-responses"

Evidence

Attach at least one:

• Failing test/log before + passing after
• Trace/log snippets
• Screenshot/recording
• Perf numbers (if relevant)

Human Verification (required)

What you personally verified (not just CI), and how:

• Verified scenarios:
  Manually installed openclaw, did the onboarding with custom provider interactive onboarding, verified tui & agent tool calls, verified the config file generated
  Response sanitization, etc not needed for azure-openai-responses as we have done for openai-responses, verified it manually, if we add the sanitization, we get 400 errors for azure-openai-responses
• Edge cases checked:
• What you did not verify:

Review Conversations

• I replied to or resolved every bot review conversation I addressed in this PR.
• I left unresolved only the conversations that still need reviewer or maintainer judgment.

If a bot review conversation is addressed by this PR, resolve that conversation yourself. Do not leave bot review conversation cleanup for maintainers.

Compatibility / Migration

• Backward compatible? Not applicable
• Config/env changes? Not applicable
• Migration needed? Not applicable
• If yes, exact upgrade steps:

Failure Recovery (if this breaks)

• How to disable/revert this change quickly: Revert the PR
• Files/config to restore:
• Known bad symptoms reviewers should watch for:

Risks and Mitigations

None

List only real risks for this PR. Add/remove entries as needed. If none, write None.

• Risk:
  • Mitigation: