release：v0.5.1

[ding113]

Summary

Release v0.5.1 - Major feature release with vendor-endpoint architecture, availability monitoring, connection pooling, and critical rate limit fixes.

Related Issues:

• Fixes bug: 5h限额设置给用户并没有生效，key并没有继承用户的5h限额，而是直接无视 #634 - User rate limit enforcement bug (5h/daily/weekly/monthly limits not enforced)
• Fixes 渠道返回 503/400 错误时，系统在原渠道重复重试而非切换至其他备用渠道 #629 - Provider failover issues (system repeatedly retrying same provider instead of switching)
• Partially addresses UI - user and admin user management issue #394 - UI/i18n improvements (logs table, group tooltips, i18n display bugs)

What's New in v0.5.1

🚀 Major Features

1. Vendor-Endpoint Architecture (#608)

• Multi-endpoint support: Configure multiple endpoints per provider for high availability
• Intelligent routing: Automatic endpoint selection based on health status and priority
• Health probing: Real-time endpoint monitoring with probe history tracking
• Circuit breakers: Dual-level protection (endpoint + vendor-type) for automatic failover
• Database migration: New tables for provider_vendors, provider_endpoints, provider_endpoint_probe_logs

2. Availability Dashboard (#646)

• Provider monitoring: Real-time availability metrics with confidence badges
• Endpoint visualization: Latency curves, probe grids, and probe terminal
• Time range selection: Flexible time windows with auto-refresh
• Overview metrics: Gauge cards for system-wide health indicators

3. Agent Pool for Connection Management (#646)

• HTTP/2 connection pooling: Reusable agents with LRU eviction
• Memory leak prevention: Proper agent lifecycle management with TTL-based expiration
• Health tracking: SSL certificate error detection with automatic agent replacement
• Race condition prevention: Concurrent request handling with proper stats tracking

4. Enhanced Logs UI (#654)

• Modular filters: Improved filtering with better UX
• Error details: Enhanced error information display
• Performance: Optimized rendering for large log datasets

🐛 Critical Fixes

User Rate Limit Enforcement (#643) - Fixes #634

Problem: User-level limits (5h/daily/weekly/monthly/concurrent sessions) were not enforced due to incorrect null-checking logic treating 0 as falsy.

Solution:

• Fixed validation logic in key actions to properly handle 0 (unlimited) vs null (not set)
• Added user concurrent session enforcement in rate limit pipeline
• Extended session tracking to support user-level counting
• Populated missing user limit fields in database queries

Impact: Keys can no longer bypass user quotas. User-level limits now properly enforced across all dimensions.

Provider Failover Logic (#633) - Fixes #629

Problem: When a provider redirected models (e.g., claude-* → glm-*), failover logic used the redirected model name, causing premature provider exhaustion.

Solution:

• Use getOriginalModel() instead of getCurrentModel() for provider selection
• Ensure failover validates against user's originally requested model
• Made response converters tolerant of unexpected tool_result blocks

Impact: Proper failover to backup providers when primary provider fails, even with model redirects.

Endpoint Stickiness on Retries (#651) - Partially addresses #629

Problem: Retry logic cycled through endpoints for both network and HTTP errors, causing suboptimal retry behavior.

Solution:

• Introduced currentEndpointIndex to track endpoint selection independently from attemptCount
• SYSTEM_ERROR (network failures): Advances to next endpoint
• PROVIDER_ERROR (HTTP 4xx/5xx): Maintains same endpoint (sticky behavior)
• No wrap-around when endpoints exhausted

Impact: Smarter retry behavior - network errors try different endpoints, HTTP errors retry same endpoint.

SSL Certificate Error Detection (#646)

• Added isSSLCertificateError() function to detect SSL/TLS errors
• Mark agents as unhealthy when SSL errors occur
• Trigger agent replacement on next request
• Limit endpoint candidates to maxRetryAttempts to prevent excessive retries

🎨 UX Improvements

Dashboard Enhancements (#650, #637)

• Token statistics modal: Detailed breakdown (input/output/cache tokens)
• Expiry countdown badge: Shows days remaining when ≤7 days
• Cache hit rate indicators: Color-coded visualization (green ≥85%, yellow ≥60%, orange <60%)
• Live sessions panel: Improved status display with concurrent count
• Chart improvements: Fixed tooltip styling, dynamic locale support

Provider Form Refactoring (#637)

• Modular architecture: Split 2215-line monolithic form into 8 focused files
• Scroll navigation: Vertical scroll with auto-updating tab indicators
• React Context + useReducer: Clean state management
• Removed deprecated types: claude-auth, gemini-cli no longer available
• Quick Paste feature: Auto-parse provider credentials from clipboard

🔧 Additional Improvements

Session Management (#646)

• Batch concurrent count query (getConcurrentCountBatch()) to avoid N+1 queries
• Session status utilities (getSessionDisplayStatus()) for consistent status calculation
• Added concurrentCount field to ActiveSessionInfo type

Thinking Signature Rectifier (#646)

• Handle "Extra inputs are not permitted" error from third-party APIs
• Support channels that don't accept signature field in thinking blocks

i18n Updates

• Comprehensive translations for availability dashboard across all 5 languages
• Network section translations for provider form (proxy settings)
• Fixed hardcoded zh-CN locale in chart components

Database Migration

Migration 0056 includes:

• New tables: provider_vendors, provider_endpoints, provider_endpoint_probe_logs
• New column: provider_vendor_id on providers table (auto-backfilled from URL domains)

To apply:

bun run db:migrate

Or enable AUTO_MIGRATE=true in environment variables.

Breaking Changes

✅ No breaking changes - All changes are backward compatible.

• Existing providers automatically migrated to vendor-endpoint structure
• API signatures remain unchanged
• Existing configurations require no manual updates

Files Changed

• 100 files modified (39,862 additions, 4,438 deletions)
• Core areas: Proxy pipeline, rate limiting, session tracking, dashboard UI, provider management
• i18n: All 5 languages updated (en, ja, ru, zh-CN, zh-TW)
• Tests: Comprehensive test coverage for new features and fixes

Testing

Pre-merge Checklist

• bun run build - Production build passes
• bun run lint - Biome check passes
• bun run typecheck - TypeScript check passes
• bun run test - All unit tests pass
• Database migration tested
• All merged PRs individually reviewed and tested

Manual Testing Recommended

1. Verify rate limit enforcement (user and key levels)
2. Test provider failover with multiple endpoints
3. Check availability dashboard displays correctly
4. Verify endpoint probe functionality
5. Test provider form in vendor view mode
6. Confirm logs UI filtering works correctly

Upgrade Notes

1. Database Migration Required: Run bun run db:migrate or enable AUTO_MIGRATE=true
2. Rate Limits Now Enforced: User-level limits will now be properly enforced. Review user limit configurations.
3. Deprecated Provider Types Removed: claude-auth and gemini-cli are no longer available in UI
4. New Features Available: Explore vendor-endpoint management in Settings > Providers (Vendor view)

Contributors

• @ding113 (26 commits)

---

Full Changelog: v0.4.3...v0.5.1

Description enhanced by Claude AI

Greptile Overview

Greptile Summary

This is a major feature release (v0.5.1) introducing vendor-endpoint architecture, availability monitoring, connection pooling, and critical bug fixes for rate limiting and provider failover.

Key Changes

Critical Fixes:

• Rate Limit Enforcement (Fix user rate limit enforcement #643): Fixed user-level limits (5h/daily/weekly/monthly/concurrent sessions) not being enforced due to incorrect null-checking that treated 0 as falsy. Now properly distinguishes between 0 (unlimited) and null (not set).
• Provider Failover (Fix provider exhaustion after model redirect (refs #629) #633): Fixed failover logic to use getOriginalModel() instead of getCurrentModel(), preventing premature provider exhaustion when model redirects occur.
• Endpoint Stickiness (fix(proxy): implement endpoint stickiness on retries #651): Implemented separate currentEndpointIndex tracking - network errors advance to next endpoint, HTTP errors retry same endpoint (sticky behavior).

Major Features:

• Vendor-Endpoint Architecture (feat(providers): add vendor endpoints with probes and failover #608): Multi-endpoint support per provider with intelligent routing, health probing, and dual-level circuit breakers (endpoint + vendor-type).
• Agent Pool (feat: multi-module improvements batch #646): HTTP/2 connection pooling with LRU eviction, TTL-based expiration, health tracking, and race condition prevention.
• Availability Dashboard (feat: multi-module improvements batch #646): Real-time provider/endpoint monitoring with latency curves, probe grids, and system health metrics.

Database Migration:

• Migration 0056 adds provider_vendors, provider_endpoints, provider_endpoint_probe_logs tables with proper indexes and foreign keys.
• Auto-backfills provider_vendor_id from existing URL domains.

UI/UX Improvements:

• Enhanced logs UI with modular filters and error details (feat(dashboard): enhance logs UI with modular filters and error details #654)
• Provider form refactoring - split 2215-line monolithic form into 8 focused files (refactor(UI): UX Improvements #637)
• Token statistics modal, expiry countdown badges, cache hit rate indicators (fix(ui): dashboard UX improvements and i18n fixes #650)
• Comprehensive i18n updates across all 5 languages

Issues Found

• Version mismatch: VERSION and package.json show 0.5.0 but PR title indicates v0.5.1 - needs correction

Confidence Score: 4/5

• Safe to merge after fixing version mismatch - changes are well-tested with comprehensive test coverage
• Score of 4 reflects solid implementation of complex features with only minor version inconsistency issue. The rate limit fix addresses a critical security bug, failover logic improvements prevent service degradation, and new architecture is well-designed with proper database migrations and comprehensive tests. Version mismatch is trivial to fix.
• VERSION and package.json need version correction to 0.5.1

Important Files Changed

Filename	Overview
VERSION	Version mismatch - file shows 0.5.0 but PR title indicates v0.5.1
package.json	Version field shows 0.5.0, inconsistent with PR title v0.5.1
src/actions/keys.ts	Fixed critical rate limit validation bug by properly checking null vs 0 values
src/app/v1/_lib/proxy/rate-limit-guard.ts	Added user-level concurrent session enforcement, completing rate limit coverage
src/app/v1/_lib/proxy/provider-selector.ts	Improved provider selection with model matching using getOriginalModel for proper failover
src/app/v1/_lib/proxy/forwarder.ts	Implemented endpoint stickiness with separate endpoint tracking for smart retry behavior
src/lib/proxy-agent/agent-pool.ts	New agent pool with HTTP/2 support, LRU eviction, health tracking, and race condition protection
drizzle/0056_tidy_quasar.sql	New vendor-endpoint architecture with 3 tables, proper indexes, and foreign key constraints

Sequence Diagram

sequenceDiagram
    participant Client
    participant ProxyForwarder
    participant RateLimitGuard
    participant ProviderSelector
    participant EndpointSelector
    participant AgentPool
    participant CircuitBreaker
    participant Provider

    Client->>ProxyForwarder: API Request
    ProxyForwarder->>RateLimitGuard: Check Rate Limits
    
    Note over RateLimitGuard: Check in order:<br/>1. Key/User Total Limits<br/>2. Key/User Concurrent Sessions<br/>3. User RPM<br/>4. Key/User 5h/Daily/Weekly/Monthly
    
    alt Rate Limit Exceeded
        RateLimitGuard-->>Client: 429 Rate Limit Error
    end
    
    RateLimitGuard->>ProviderSelector: Select Provider
    ProviderSelector->>ProviderSelector: Filter by Group/Model/Format
    ProviderSelector->>CircuitBreaker: Check Circuit State
    
    alt Circuit Open
        ProviderSelector->>ProviderSelector: Exclude Provider
    end
    
    ProviderSelector->>ProviderSelector: Select by Priority/Weight
    ProviderSelector-->>ProxyForwarder: Selected Provider
    
    ProxyForwarder->>EndpointSelector: Get Endpoints for Vendor
    EndpointSelector-->>ProxyForwarder: Endpoint List (sorted by latency)
    
    loop Retry Logic (Endpoint Stickiness)
        ProxyForwarder->>AgentPool: Get/Create Agent
        
        alt Agent Unhealthy
            AgentPool->>AgentPool: Evict & Create New
        end
        
        AgentPool-->>ProxyForwarder: HTTP Agent
        
        ProxyForwarder->>Provider: Forward Request via Endpoint
        
        alt Network Error (SYSTEM_ERROR)
            Provider-->>ProxyForwarder: Network Failure
            ProxyForwarder->>CircuitBreaker: Record Endpoint Failure
            ProxyForwarder->>ProxyForwarder: Advance to Next Endpoint
        else HTTP Error (PROVIDER_ERROR)
            Provider-->>ProxyForwarder: HTTP 4xx/5xx
            ProxyForwarder->>ProxyForwarder: Retry Same Endpoint
        else Success
            Provider-->>ProxyForwarder: Success Response
            ProxyForwarder->>CircuitBreaker: Record Endpoint Success
        end
    end
    
    alt All Endpoints Failed
        ProxyForwarder->>ProviderSelector: Select Next Provider
        ProviderSelector-->>ProxyForwarder: Fallback Provider
    else All Providers Failed
        ProxyForwarder-->>Client: 503 No Available Providers
    else Request Succeeded
        ProxyForwarder-->>Client: Success Response
    end

Loading

[coderabbitai]

Important

Review skipped

Too many files!

150 files out of 300 files are above the max files limit of 150.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Note

Other AI code review bot(s) detected

CodeRabbit has detected other AI code review bot(s) in this pull request and will avoid duplicating their findings in the review comments. This may lead to a less comprehensive review.

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch dev

---

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.}

[gemini-code-assist]

Summary of Changes

Hello @ding113, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request introduces a comprehensive set of features and improvements focused on enhancing observability, management, and user experience within the system. Key areas of focus include real-time provider health monitoring, a redesigned dashboard with richer metrics and active session insights, and a more detailed and organized view for request logs. The changes also streamline provider configuration and management through new UI elements and batch operations.

Highlights

• Provider Endpoint Probing: Introduced a new feature for real-time monitoring of provider endpoint availability and performance. This includes new database tables (provider_endpoint_probe_logs, provider_endpoints, provider_vendors) and related configuration options in .env.example.
• Dashboard Enhancements: The dashboard now features a new Bento Grid layout, including metric cards, live session panels, and leaderboard cards. The statistics chart has been updated with new functionalities like chart mode toggles (stacked/overlay) and user filtering.
• Usage Log Details Improvement: The request details dialog in usage logs has been refactored into a tabbed interface (Summary, Logic Trace, Performance, Metadata) for better organization and presentation of information.
• Provider Management UI Overhaul: Significant updates to the provider settings UI, introducing a 'Vendor' view, batch editing capabilities, and a 'Quick Paste' feature for easier provider configuration.
• Internationalization Updates: Extensive additions and modifications to English, Japanese, Russian, Simplified Chinese, and Traditional Chinese message files to support the new features and UI changes.
• Version Bump: The project version has been updated from 0.4.3 to 0.5.0.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[github-actions]

Code Review Summary

This is a massive release PR (v0.5.1) merging dev into main with significant changes across the codebase. After thorough analysis, I've identified one CRITICAL issue that must be addressed before merging.

PR Size: XL

• Lines changed: 44,300 (39,862 additions + 4,438 deletions)
• Files changed: 350

⚠️ CRITICAL: This PR is extremely large and should be split into smaller, reviewable chunks. A PR of this size:

• Makes thorough code review nearly impossible
• Increases risk of introducing bugs
• Makes it difficult to identify the root cause if issues arise
• Violates best practices for incremental delivery

Recommendation: Future releases should be broken down into feature-specific PRs (e.g., agent pool implementation, availability dashboard, provider endpoints refactor, etc.) that can be reviewed and merged incrementally.

---

Issues Found

Category	Critical	High	Medium	Low
Standards Violation	1	0	0	0
Error Handling	0	0	1	0
Code Quality	0	0	1	0

---

Critical Issues (Must Fix)

1. [CRITICAL] Wrong PR Target Branch (CLAUDE.md violation)

Location: PR configuration
Issue: This PR targets main branch, but CLAUDE.md explicitly requires all PRs to target dev branch.

Evidence from CLAUDE.md:

PR Target Branch: dev (all pull requests must target the dev branch)

Impact: Bypasses the intended branching strategy, potentially skipping integration testing and staging validation.

Fix Required: Close this PR and recreate it targeting dev branch instead of main.

Confidence Score: 100/100

---

Medium Priority Issues (Consider Fixing)

1. Client-Side Error Logging Pattern

Locations: Multiple client components use console.error instead of proper error handling:

• src/app/[locale]/dashboard/availability/_components/availability-dashboard.tsx:64
• src/app/[locale]/dashboard/logs/_components/error-details-dialog/index.tsx:116
• src/app/[locale]/settings/providers/_components/forms/provider-form/index.tsx (multiple instances)
• src/app/api/availability/endpoints/route.ts:42

Issue: While console.error is acceptable for client-side debugging, the API route at src/app/api/availability/endpoints/route.ts:42 should use the centralized logger for consistency and proper log aggregation.

Suggested fix for API route:

// Instead of:
console.error("Endpoint availability API error:", error);

// Use:
import { logger } from "@/lib/logger";
logger.error("Endpoint availability API error", { error });

Confidence Score: 85/100

2. Type Safety - Intentional any Usage

Locations:

• src/repository/provider-endpoints.ts (lines with toProviderVendor, toProviderEndpoint, toProviderEndpointProbeLog)
• src/repository/_shared/transformers.ts

Observation: These files use // eslint-disable-next-line @typescript-eslint/no-explicit-any with any types for database row transformers.

Assessment: This appears to be intentional design for handling dynamic database query results. The transformer functions validate and type-cast the data appropriately. While not ideal, this is a pragmatic approach for ORM result handling.

Recommendation: Consider using Drizzle's typed query results if possible, but current implementation is acceptable given the validation logic in place.

Confidence Score: 80/100 (intentional design, not a defect)

---

Positive Observations

✅ Excellent test coverage: New features like agent-pool have comprehensive unit tests (467 lines of tests for 450 lines of implementation)
✅ Good error handling: The AgentPoolImpl properly handles agent cleanup, TTL expiration, and LRU eviction
✅ Type safety: Most new code uses proper TypeScript types with minimal any usage
✅ Internationalization: All user-facing strings properly use i18n across 5 languages
✅ Documentation: Complex modules like agent-pool.ts have clear JSDoc comments explaining purpose and behavior

---

Review Coverage

• Logic and correctness - Reviewed
• Security (OWASP Top 10) - No issues found
• Error handling - Minor improvements suggested
• Type safety - Acceptable with intentional any usage
• Documentation accuracy - Good
• Test coverage - Excellent for new features
• Code clarity - Good
• Standards compliance - CRITICAL violation found

---

Action Required

CRITICAL: This PR must be closed and recreated targeting the dev branch per CLAUDE.md requirements. The main branch should only receive merges from dev after proper validation.

---

Automated review by Claude AI

[gemini-code-assist]

Code Review

此 Pull Request 为 v0.5.0 版本引入了大量新功能和改进，包括一个全新设计的仪表盘（采用 bento 网格布局）、一个新的供应商可用性监控页面，以及大幅增强的用量日志筛选和详情视图。数据库 schema 已更新，以支持新的供应商/服务商/端点模型。代码质量总体较高，很好地运用了现代 React hooks 和服务器操作。

我发现了一个数据库迁移中的关键问题，可能会影响现有部署。此外，还存在一个版本号不一致的小问题和一个可能导致混淆的环境变量配置。请查看具体评论以获取建议。

[gemini-code-assist]

ALTER TABLE "providers" ADD COLUMN IF NOT EXISTS "provider_vendor_id" integer; 语句添加了一个可为空的整数列。然而，相应的快照文件 drizzle/meta/0056_snapshot.json 将此列定义为 notNull: true。对于已有数据的表，这将导致新列的值为 NULL，与应用程序的 schema 定义（Drizzle schema object）冲突，很可能在运行时引发错误。

为了确保迁移的安全性，建议分步操作：

1. 添加可为空的列。
2. 编写一个数据迁移脚本，为 providers 表中的现有行填充 provider_vendor_id 的值。
3. 添加 NOT NULL 约束。

[gemini-code-assist]

PROBE_INTERVAL_MS 和 PROBE_TIMEOUT_MS 这两个变量似乎已被新的 ENDPOINT_PROBE_* 变量所取代。如果它们已不再使用，建议将其移除以避免配置混淆。如果它们仍用于其他目的，最好添加注释说明其具体用途。

[gemini-code-assist]

此文件中的版本号已更新为 0.5.0，package.json 中也是如此，但 Pull Request 的标题是 v0.5.1。请确保版本号在各处保持一致，以避免发布混淆。

[gemini-code-assist]

版本号已更新为 0.5.0，但 Pull Request 的标题是 v0.5.1。请确保 PR 标题、VERSION 文件和 package.json 中的版本号保持一致。

[greptile-apps]

_{1 file reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

VERSION file shows 0.5.0 but PR title says v0.5.1 - check which version is correct and ensure consistency

Prompt To Fix With AI

This is a comment left during a code review.
Path: VERSION
Line: 1:1

Comment:
VERSION file shows `0.5.0` but PR title says `v0.5.1` - check which version is correct and ensure consistency

How can I resolve this? If you propose a fix, please make it concise.

[greptile-apps]

_{2 files reviewed, 2 comments}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

Version file shows 0.5.0 but PR title indicates v0.5.1

Suggested change

	0.5.0
	0.5.1

Prompt To Fix With AI

This is a comment left during a code review.
Path: VERSION
Line: 1:1

Comment:
Version file shows `0.5.0` but PR title indicates `v0.5.1`

```suggestion
0.5.1
```

How can I resolve this? If you propose a fix, please make it concise.

[greptile-apps]

package.json version should match the VERSION file and PR title

Suggested change

	"version": "0.5.0",
	"version": "0.5.1",

Prompt To Fix With AI

This is a comment left during a code review.
Path: package.json
Line: 3:3

Comment:
`package.json` version should match the VERSION file and PR title

```suggestion
  "version": "0.5.1",
```

How can I resolve this? If you propose a fix, please make it concise.

[greptile-apps]

_{2 files reviewed, 2 comments}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

Version should be 0.5.1 to match PR title

Suggested change

	0.5.0
	0.5.1

Prompt To Fix With AI

This is a comment left during a code review.
Path: VERSION
Line: 1:1

Comment:
Version should be `0.5.1` to match PR title

```suggestion
0.5.1
```

How can I resolve this? If you propose a fix, please make it concise.

[greptile-apps]

Version should be 0.5.1 to match PR title and VERSION file

Suggested change

	"version": "0.5.0",
	"version": "0.5.1",

Prompt To Fix With AI

This is a comment left during a code review.
Path: package.json
Line: 3:3

Comment:
Version should be `0.5.1` to match PR title and VERSION file

```suggestion
  "version": "0.5.1",
```

How can I resolve this? If you propose a fix, please make it concise.