mirror of
https://github.com/justlovemaki/AIClient-2-API.git
synced 2026-04-25 09:25:59 +03:00
[PR #74] [CLOSED] This PR extends AIClient-2-API with production-ready support for Warp AI and Ollama, implementing a unified routing system that enables seamless multi-provider integration through a single base URL with intelligent model prefixes. #256
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/AIClient-2-API-justlovemaki#256
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
📋 Pull Request Information
Original PR: https://github.com/justlovemaki/AIClient-2-API/pull/74
Author: @HALDRO
Created: 11/14/2025
Status: ❌ Closed
Base:
main← Head:main📝 Commits (4)
d6c620dfeat: add enterprise-grade Warp and Ollama protocol support with advanced routingd3aef5bfeat: Add model prefix system and parallel model fetching7d50378docs: Update README and translate scripts to English3f311d8fix: correct provider selection by using model prefix before removal📊 Changes
42 files changed (+5934 additions, -248 deletions)
View changed files
📝
.gitignore(+3 -1)📝
README.md(+87 -18)📝
config.json.example(+5 -1)📝
install-and-run.bat(+39 -39)📝
install-and-run.sh(+44 -44)📝
package-lock.json(+96 -3)📝
package.json(+1 -0)➕
proto/attachment.proto(+57 -0)➕
proto/citations.proto(+20 -0)➕
proto/debug.proto(+12 -0)➕
proto/file_content.proto(+18 -0)➕
proto/input_context.proto(+64 -0)➕
proto/options.proto(+12 -0)➕
proto/request.proto(+173 -0)➕
proto/response.proto(+159 -0)➕
proto/suggestions.proto(+22 -0)➕
proto/task.proto(+503 -0)➕
proto/todo.proto(+23 -0)📝
run-docker.bat(+22 -22)📝
run-docker.sh(+22 -22)...and 22 more files
📄 Description
Many tools (e.g., Copilot Chat non-Insider) don't fully support OpenAI protocol. This implementation provides bidirectional protocol conversion and flexible routing to solve compatibility issues.
✨ Major Features
d6c620d)✅ Bidirectional converter: Full OpenAI ↔ Ollama protocol translation
✅ Streaming support: Real-time response streaming with proper chunk formatting
✅ Native integration: OllamaConverter strategy following existing architecture patterns
✅ Use case: Enables tools with limited OpenAI support (like Copilot Chat) to work with Ollama models
Technical implementation:
New converter: src/converters/strategies/OllamaConverter.js
Handles request/response transformation between Ollama and OpenAI formats
Supports both streaming and non-streaming modes
Proper UUID and timestamp generation for Ollama responses
2. Warp AI Integration (
d6c620d)✅ Complete Warp module ecosystem (10 specialized handlers):
Authentication & session management
Protocol Buffers support (11 proto definitions)
Model discovery and management
Request/response packet building
Advanced reordering logic
✅ Enterprise-grade architecture: Production-ready with proper error handling
⚠️ Note: Warp lacks official API documentation and may require future adjustments if Warp tightens restrictions. Consider using Warp CLI for improved stability.
Technical implementation:
Full module: src/warp/ (10 files)
Proto definitions: proto/warp/ (11 .proto files)
Strategy: src/warp/warp-strategy.js
Adapter: Extended in src/adapter.js
3. Unified Model Prefix System (
d3aef5b)✅ Single Base URL: All providers accessible through one endpoint
✅ Intelligent routing: Automatic provider detection via model prefixes
[Warp] gpt-5 → routes to Warp
[Kiro] claude-sonnet-4.5 → routes to Kiro
✅ Parallel model fetching: Improved performance for model list retrieval
✅ Backward compatible: Existing configurations work without changes
Technical implementation:
Enhanced MODEL_PREFIX_MAP in src/common.js
Updated model-provider-mapper.js with prefix stripping logic
Parallel async model fetching in UI manager
Automatic prefix removal before API calls
4. Documentation & Localization (
7d50378)✅ Script updates: install-and-run.bat/sh, run-docker.bat/sh
🏗️ Architecture Changes
New files:
src/converters/strategies/OllamaConverter.js # Ollama protocol converter
src/warp/ # Complete Warp module (10 files)
proto/warp/ # Warp Protocol Buffers (11 files)
Modified files:
src/common.js # Model prefix system, provider constants
src/model-provider-mapper.js # Prefix-aware routing logic
src/ollama-handler.js # Enhanced Ollama request handling
src/ui-manager.js # Parallel model fetching
src/adapter.js # Warp adapter integration
README.md # Documentation updates
🎯 Benefits
Greater flexibility: Single base URL with prefix-based routing simplifies client configuration
Broader compatibility: Ollama converter solves OpenAI protocol limitations in various tools
Future-proof: Modular architecture makes adding new providers straightforward
Performance: Parallel model fetching reduces UI load times
User experience: Clear model prefixes improve provider visibility
✅ Testing
Only basic manual tests with simple queries were performed, and the basic tools worked.
[x] Ollama ↔ OpenAI conversion tested with streaming
[x] Warp API integration verified with authentication flow
[x] Model prefix routing tested across all providers
[x] Parallel model fetching validated
[x] Backward compatibility confirmed with existing configs
⚠️ Known Limitations
Warp Integration:
No official API documentation available
Configuration complexity due to undocumented endpoints
May require updates if Warp implements API restrictions
Recommendation: Consider Warp CLI integration for improved stability
📦 Commits
d6c620d- feat: add enterprise-grade Warp and Ollama protocol support with advanced routingd3aef5b- feat: Add model prefix system and parallel model fetching7d50378- docs: Update README and translate scripts to EnglishThis PR contains multiple independent improvements. I'm not planning to split it into separate PRs - feel free to cherry-pick individual commits that align with your project goals. Each commit is self-contained and can be merged independently.
I hope you'll support this contribution to make AIClient-2-API even more versatile! 🚀
Example from Copilot Chat integrated.
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.