PR #62155: fix(kimi-coding): change default API from anthropic-messages to openai-completions

• Repository: openclaw/openclaw
• Author: demrane
• State: open | merged: False
• Link: https://github.com/openclaw/openclaw/pull/62155

Description (problem / solution / changelog)

Summary

• Problem: Kimi API is OpenAI-compatible, not Anthropic-compatible. The previous default caused tool calls to fail with malformed output (e.g. raw <|tool_calls_section_begin|> markers leaking to users).
• Why it matters: Tool calls were broken for Kimi-coding users, showing raw markers instead of proper tool execution.
• What changed: Changed API type from "anthropic-messages" to "openai-completions", updated baseUrl to include /v1/, updated all related tests.
• What did NOT change (scope boundary): No changes to other providers or core agent logic.

Change Type (select all)

• Bug fix
• Feature
• Refactor required for the fix
• 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 #61379
• Related #
• This PR fixes a bug or regression

Root Cause (if applicable)

For bug fixes or regressions, explain why this happened, not just what changed. Otherwise write N/A. If the cause is unclear, write Unknown.

• Root cause: Wrong API type configured - Kimi uses OpenAI-compatible API, not Anthropic.
• Missing detection / guardrail: N/A
• Contributing context (if known): API compatibility mismatch.

Regression Test Plan (if applicable)

For bug fixes or regressions, name the smallest reliable test coverage that should catch this. Otherwise write N/A.

• Coverage level that should have caught this:
  • Unit test
  • Seam / integration test
  • End-to-end test
  • Existing coverage already sufficient
• Target test or file: kimi-coding extension tests
• Scenario the test should lock in: Tool calls work correctly with OpenAI-compatible API.
• Why this is the smallest reliable guardrail: Tests verify the API configuration.
• Existing test that already covers this (if any): All 10 kimi-coding tests pass.
• If no new test is added, why not: N/A - configuration fix.

User-visible / Behavior Changes

List user-visible changes (including defaults/config). If none, write None.

Tool calls now work correctly with Kimi-coding instead of showing raw markers.

Diagram (if applicable)

For UI changes or non-trivial logic flows, include a small ASCII diagram reviewers can scan quickly. Otherwise write N/A.

N/A

Security Impact (required)

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

Repro + Verification

Environment

• OS: Linux
• Runtime/container: Node.js
• Model/provider: Kimi K2.5
• Integration/channel (if any): N/A
• Relevant config (redacted): N/A

Steps

1. Use kimi-coding model
2. Execute a tool call
3. Verify no raw markers leak to user

Expected

Tool calls execute cleanly.

Actual

Fixed - tool calls work correctly.

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: Tool calls work with Kimi
• Edge cases checked: N/A
• What you did not verify: N/A

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? (Yes - fixes broken behavior)
• Config/env changes? (No - internal default change)
• Migration needed? (No)
• If yes, exact upgrade steps:

Risks and Mitigations

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

None

Changed files

• extensions/kimi-coding/implicit-provider.test.ts (modified, +1/-1)
• extensions/kimi-coding/onboard.ts (modified, +1/-1)
• extensions/kimi-coding/provider-catalog.test.ts (modified, +2/-2)
• extensions/kimi-coding/provider-catalog.ts (modified, +2/-2)
• extensions/kimi-coding/stream.test.ts (modified, +5/-5)
• src/agents/pi-embedded-runner.anthropic-tool-replay.live.test.ts (modified, +1/-1)
• src/plugins/loader.test.ts (modified, +2/-2)
• src/plugins/loader.ts (modified, +2/-2)