feat(llm): add ZhipuAIProvider with logging support

[baleen37]

Summary

• Add ZhipuAIProvider implementation using zhipuai-sdk-nodejs-v4 SDK
• Add comprehensive logging to ZhipuAIProvider.complete() method
• Add comprehensive logging to batch-extract-prompt.extractObservationsFromBatch()
• Add Zhipu AI setup documentation (docs/zhipu-ai-setup.md)
• Require provider field in LLMConfig configuration

Changes

• ZhipuAIProvider: New LLM provider supporting GLM-4.x models (glm-4.7 default)
• Logging: Added structured logging with duration tracking for both providers and batch extraction
• Documentation: Complete setup guide for Zhipu AI API key and configuration
• Config: Updated LLMConfig to require explicit provider field

Test Plan

• All LLM unit tests pass (79 tests)
• batch-extract-prompt tests pass (16 tests)
• TypeScript typecheck passes for modified files
• Documentation verified and markdownlint validated

🤖 Generated with Claude Code

Summary by CodeRabbit

• New Features

  • Added support for Zhipu AI as a conversation summarization provider alongside Gemini
  • Configuration schema now supports multiple providers with provider-agnostic setup

• Documentation

  • Added comprehensive setup guides and configuration examples for Zhipu AI integration
  • Updated provider configuration documentation to reflect multi-provider support

• Tests

  • Added test coverage for Zhipu AI provider functionality
  • Updated existing tests for async provider initialization

[coderabbitai]

Caution

Review failed

The pull request is closed.

📝 Walkthrough

Walkthrough

The pull request adds multi-provider LLM support to the conversation-memory plugin by introducing Zhipu AI as an alternative to Gemini. The createProvider function converts to async with dynamic provider selection, a new ZhipuAIProvider class is implemented, comprehensive setup documentation and configuration examples are added, and the zhipuai-sdk-nodejs-v4 dependency is included.

Changes

Cohort / File(s)	Summary
Documentation and Setup README.md, docs/zhipu-ai-setup.md, skills/configure/SKILL.md	Updated README to support multiple providers with provider-agnostic schema; added comprehensive Zhipu AI setup guide covering API key acquisition, configuration, debugging, and troubleshooting; added SKILL.md documenting configure skill with multi-provider configuration and environment variables.
Configuration Examples skills/configure/examples/gemini-config.json, skills/configure/examples/minimal-config.json, skills/configure/examples/zhipu-ai-config.json, skills/configure/examples/exclude.txt	Added example configuration files for Gemini and Zhipu AI providers with placeholder API keys and models; added exclude.txt template for project filtering.
Package Dependencies package.json	Added zhipuai-sdk-nodejs-v4 (^0.1.12) runtime dependency to support Zhipu AI provider integration.
LLM Core Infrastructure src/core/llm/config.ts, src/core/llm/config.test.ts	Introduced LLMProviderType type supporting 'gemini' and 'zhipu-ai'; converted createProvider from synchronous to async with dynamic imports; added provider-specific default models; enhanced validation for provider values and required fields; updated tests to verify async behavior and provider-aware configurations.
Provider Implementations src/core/llm/gemini-provider.ts, src/core/llm/gemini-provider.test.ts, src/core/llm/zhipu-provider.ts, src/core/llm/zhipu-provider.test.ts	Added ZhipuAIProvider class with completion, logging, and token usage tracking; added comprehensive unit tests for Zhipu provider with mocked API; simplified Gemini provider tests to remove internal mock assertions; enhanced Gemini provider with structured logging and timing.
LLM Index Exports src/core/llm/index.ts, src/core/llm/index.test.ts	Exported new ZhipuAIProvider class; updated LLMConfig type to use flat apiKey field instead of nested gemini-specific config; updated tests to reflect new config shape.
Integration and Batch Processing src/core/llm/batch-extract-prompt.ts, src/core/summarizer.ts, test-integration.ts, test-roundrobin.ts	Enhanced batch extraction with structured logging (start/end timing, event counts, prompt length, token usage); converted provider creation to async with await; updated integration tests to accept both 'gemini' and 'zhipu-ai' providers and use generic config fields; updated test-roundrobin to remove unused imports and await provider creation.
Test File test.txt	Added minimal test placeholder file.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• feat(conversation-memory): restore conversation-memory plugin #425: Directly extends conversation-memory plugin with multi-provider LLM support, overlapping the same core files and functionality.
• fix(conversation-memory): move vitest.config.ts to plugin directory to reduce resource usage #462: Modifies conversation-memory plugin at code level, overlapping in config.ts, zhipu-provider implementation, and LLM exports.
• refactor: migrate conversation-memory to me plugin #423: Operates on the same conversation-memory plugin codebase with conflicting scope (removes vs. extends plugin).

Poem

🐰 A hop through providers so fine,
Gemini's great, but Zhipu will shine!
Async awaits the LLM's call,
Multi-provider support for all—
Logs and configs in perfect line,
This plugin's future? Oh, quite divine! ✨

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch ralph/second-feature

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}