vxcontrol/pentagi
1
0
Fork 0
mirror of https://github.com/vxcontrol/pentagi.git synced 2026-07-21 08:15:23 -04:00
Code Issues 45 Packages Projects Releases 9 Wiki Activity

[PR #102] [MERGED] langchaingo update and observability improvements #161

New Issue
Closed
opened 2026-06-06 22:09:28 -04:00 by yindo · 0 comments
yindo commented 2026-06-06 22:09:28 -04:00
Owner
Copy Link
Copy Source

📋 Pull Request Information

Original PR: https://github.com/vxcontrol/pentagi/pull/102
Author: @asdek
Created: 2/18/2026
Status: ✅ Merged
Merged: 2/18/2026
Merged by: @asdek

Base: master ← Head: feat/langchaingo_update

📝 Commits (10+)

  • e6cbf4a feat: major langchaingo dependency update and add product working analysis API
  • 8f1f9ee feat: update graphiti-go-client dependency and enhance observability integration
  • fd60a8f feat: add analytics service and endpoints for usage statistics
  • e193390 fix: typo in graphiti source description
  • d307fa1 feat: enhance summarization handling for reasoning signatures
  • e68e237 feat: enhance summarization functionality for multi-provider compatibility
  • efd966c refactor: replace newID() with newSpanID() for W3C compliance
  • 16bd43c fix: chain summarization with critical guarantees
  • 4065300 chore: update base image in Dockerfile from alpine:3.22.1 to alpine:3.23.3
  • d84d93b chore: update model and pricing in moonshot provider configuration

📊 Changes

277 files changed (+99895 additions, -18833 deletions)

View changed files

📝 .env.example (+2 -0)
📝 .vscode/launch.json (+5 -3)
📝 .vscode/settings.json (+2 -2)
📝 Dockerfile (+8 -7)
📝 README.md (+104 -57)
📝 backend/cmd/ctester/main.go (+4 -0)
📝 backend/cmd/ftester/mocks/logs.go (+17 -1)
📝 backend/cmd/ftester/worker/executor.go (+6 -0)
📝 backend/cmd/ftester/worker/tester.go (+6 -5)
📝 backend/cmd/installer/wizard/controller/controller.go (+19 -3)
📝 backend/cmd/installer/wizard/locale/locale.go (+9 -0)
📝 backend/cmd/installer/wizard/models/llm_provider_form.go (+50 -0)
➕ backend/docs/analytics_api.md (+1005 -0)
📝 backend/docs/chain_ast.md (+106 -1)
📝 backend/docs/chain_summary.md (+208 -10)
📝 backend/docs/config.md (+158 -144)
📝 backend/docs/controller.md (+1 -1)
📝 backend/docs/database.md (+587 -47)
📝 backend/docs/langfuse.md (+534 -37)
➕ backend/docs/llms_how_to.md (+1865 -0)

...and 80 more files

📄 Description

Description of the Change

This PR introduces a major update to the langchaingo dependency and implements comprehensive observability improvements across the system. It also adds analytics capabilities, enhances chain summarization logic, and includes various UI/UX improvements.

Problem

The system required several critical improvements:

  1. Outdated langchaingo dependency limiting provider compatibility and features
  2. Inconsistent observability and tracing across components
  3. Lack of system-wide analytics and usage statistics
  4. Chain summarization issues with multi-provider support (Gemini, Anthropic, Kimi/Moonshot)
  5. Missing W3C Trace Context compliance for distributed tracing
  6. Various UI/UX issues affecting user experience

Solution

Backend Infrastructure:

  • Updated langchaingo to the latest version with enhanced provider support
  • Standardized observability implementation across all components with W3C-compliant trace/span IDs
  • Introduced comprehensive analytics service with endpoints for system-wide and flow-specific usage statistics
  • Enhanced chain summarization algorithm with critical guarantees for reasoning preservation across different LLM providers
  • Added support for LLM_SERVER_PROVIDER and LLM_SERVER_PRESERVE_REASONING environment variables
  • Updated Langfuse API client with new endpoints (annotation queues, blob storage integrations, LLM connections, etc.)
  • Improved tool call ID generation and validation with function name support

Frontend Improvements:

  • Added PDF generation capabilities using @react-pdf/renderer for flow reports
  • Enhanced theme handling with proper system theme detection and persistence
  • Improved authentication flow with safe return URL handling to prevent open redirects
  • Better terminal component theme integration

DevOps & Configuration:

  • Updated base Alpine image from 3.22.1 to 3.23.3
  • Enhanced provider configurations (Moonshot model updates, additional provider files)
  • Improved VSCode debug and settings configurations

Database & Analytics:

  • Added new analytics models and queries for comprehensive usage tracking
  • Implemented tracking fields for better observability (tool calls, token usage, costs)
  • Created migration scripts for new analytics tables

Closes #

Type of Change

  • 🐛 Bug fix (non-breaking change which fixes an issue)
  • 🚀 New feature (non-breaking change which adds functionality)
  • 💥 Breaking change (fix or feature that would cause existing functionality to not work as expected)
  • 📚 Documentation update
  • 🔧 Configuration change
  • 🧪 Test update
  • 🛡️ Security update

Areas Affected

  • Core Services (Frontend UI/Backend API)
  • AI Agents (Researcher/Developer/Executor)
  • Security Tools Integration
  • Memory System (Vector Store/Knowledge Base)
  • Monitoring Stack (Grafana/OpenTelemetry)
  • Analytics Platform (Langfuse)
  • External Integrations (LLM/Search APIs)
  • Documentation
  • Infrastructure/DevOps

Testing and Verification

The changes have been tested across multiple LLM providers and system components to ensure compatibility and stability.

Test Configuration

PentAGI Version: dev (feat/langchaingo_update branch)
Docker Version: 24.x+
Host OS: Linux/macOS
LLM Provider: OpenAI, Anthropic, Google Gemini, Moonshot
Enabled Features: [Langfuse, Grafana, Analytics]

Test Steps

  1. Updated all Go dependencies and verified compilation
  2. Ran comprehensive test suites for:
    • Chain summarization with different providers (Anthropic, Gemini, Moonshot)
    • Tool call ID generation and validation
    • Analytics service endpoints
    • Observability integration (Langfuse traces, spans, events)
  3. Tested frontend changes:
    • PDF report generation from flow results
    • Theme switching (light/dark/system)
    • Authentication flow with return URL handling
  4. Verified database migrations execute successfully
  5. Tested LLM provider compatibility with new langchaingo version
  6. Validated W3C Trace Context compliance in distributed tracing
  7. Confirmed analytics endpoints return correct usage statistics

Test Results

  • All existing unit tests pass
  • New test cases added for:
    • Chain summarization with reasoning signatures
    • Analytics data conversion
    • Tool call ID pattern generation
    • Langfuse observation conversion
  • Manual testing confirmed PDF generation works correctly
  • Theme handling properly respects system preferences
  • Authentication redirects work securely without open redirect vulnerabilities
  • Analytics API returns accurate system-wide and flow-specific metrics

Security Considerations

Enhanced Security Measures:

  • Implemented getSafeReturnUrl utility to prevent open redirect vulnerabilities in authentication flow
  • All return URLs are validated to ensure they are relative paths only
  • Updated Alpine base image to 3.23.3 to include latest security patches

Dependency Updates:

  • Updated langchaingo dependency with latest security fixes
  • Updated Langfuse API client with improved error handling and security features
  • All new dependencies have been reviewed for known vulnerabilities

No Breaking Security Changes:

  • Existing authentication and authorization mechanisms remain intact
  • Session handling unchanged
  • API token management unaffected by observability changes

Performance Impact

Improvements:

  • Enhanced chain summarization algorithm is now idempotent and more efficient
  • Memoized theme detection function reduces unnecessary re-renders in terminal component
  • Optimized observability event batching in Langfuse integration
  • Better memory management in tool call ID generation

New Analytics Overhead:

  • Analytics service adds minimal overhead (<1% CPU) for usage tracking
  • Database queries optimized with proper indexing (existing tracking fields)
  • Analytics endpoints use efficient aggregation queries

Testing:

  • Performance tested with large message chains (1000+ messages)
  • Verified summarization maintains sub-second latency for typical flows
  • Analytics queries tested with 10K+ flow records
  • No measurable impact on LLM request/response latency

Recommendations:

  • Analytics data should be periodically archived for long-running systems
  • Consider adding database indices on created_at fields for faster analytics queries at scale

Documentation Updates

  • README.md updates (new LLM_SERVER_PROVIDER and LLM_SERVER_PRESERVE_REASONING variables)
  • API documentation updates (Swagger/OpenAPI docs for analytics endpoints)
  • Configuration documentation updates (backend/docs/config.md)
  • GraphQL schema updates (analytics-related queries and types)
  • Other:
    • Added backend/docs/analytics_api.md - comprehensive analytics API documentation
    • Added backend/docs/llms_how_to.md - LLM provider configuration guide
    • Updated backend/docs/observability.md - enhanced observability documentation
    • Updated backend/docs/chain_summary.md - chain summarization algorithm details
    • Updated backend/docs/chain_ast.md - AST handling for reasoning signatures
    • Updated .env.example with new configuration options

Deployment Notes

New Environment Variables:

# Optional: Specify server-side LLM provider for specific features
LLM_SERVER_PROVIDER=openai  # or anthropic, gemini, etc.

# Optional: Preserve reasoning content for providers that support it (Kimi/Moonshot)
LLM_SERVER_PRESERVE_REASONING=true

Database Migrations:
The following migrations will run automatically on deployment:

  • 20260128_153000_tool_call_id_template.sql - Adds tool call ID template tracking
  • 20260129_120000_add_tracking_fields.sql - Adds analytics tracking fields

Docker Image Updates:

  • Base Alpine image updated to 3.23.3
  • Additional provider configuration files included in the image

Recommended Deployment Steps:

  1. Pull latest changes from feat/langchaingo_update branch
  2. Update .env file with new optional variables (if needed)
  3. Run docker-compose pull to get updated base images
  4. Run docker-compose down && docker-compose up -d for clean restart
  5. Verify database migrations completed successfully (check logs)
  6. Test analytics endpoints: GET /api/v1/analytics/usage
  7. Verify Langfuse integration still working (if enabled)

Backward Compatibility:

  • All changes are backward compatible
  • Existing flows and data remain unaffected
  • New environment variables are optional with sensible defaults
  • GraphQL schema changes are additive only

Rollback Plan:
If issues occur, rollback is straightforward:

  1. Checkout previous commit
  2. Rebuild containers: docker-compose up -d --build
  3. Database migrations are forward-compatible (no rollback needed)

Checklist

Code Quality

  • My code follows the project's coding standards
  • I have added/updated necessary documentation
  • I have added tests to cover my changes
  • All new and existing tests pass
  • I have run go fmt and go vet (for Go code)
  • I have run npm run lint (for TypeScript/JavaScript code)

Security

  • I have considered security implications
  • Changes maintain or improve the security model
  • Sensitive information has been properly handled

Compatibility

  • Changes are backward compatible
  • Breaking changes are clearly marked and documented
  • Dependencies are properly updated

Documentation

  • Documentation is clear and complete
  • Comments are added for non-obvious code
  • API changes are documented

Additional Notes

Key Changes Summary:

  1. Langchaingo Update (16 commits worth of work):

    • Major dependency update bringing compatibility with latest LLM providers
    • Enhanced provider support for Anthropic, OpenAI, Gemini, Moonshot/Kimi
    • Improved reasoning signature handling across different providers

  2. Observability Enhancements:

    • W3C Trace Context compliance for distributed tracing
    • Standardized span/trace ID handling throughout the codebase
    • Enhanced Langfuse integration with new observation types (agent, chain, tool, retriever, embedding, evaluator, guardrail)
    • Added 30+ new Langfuse API client methods

  3. Analytics Infrastructure:

    • New comprehensive analytics service with system-wide and per-flow statistics
    • Tracking for token usage, costs, tool calls, and execution times
    • RESTful API endpoints for analytics data retrieval

  4. Chain Summarization Improvements:

    • Critical guarantees for preserving last N QA sections
    • Idempotent operation - multiple summarizations don't alter already summarized content
    • Provider-specific handling for reasoning signatures (Gemini, Anthropic, Kimi)
    • Enhanced testing coverage for edge cases

  5. Frontend Enhancements:

    • PDF export functionality for flow reports using @react-pdf/renderer
    • Improved theme system with proper system theme detection
    • Secure authentication flow preventing open redirects

Review Focus Areas:

  • Chain summarization logic (complex but critical for multi-provider support)
  • Analytics service database queries and performance
  • Observability integration changes (verify Langfuse still works as expected)
  • PDF generation quality and formatting
  • Security of authentication return URL handling

Testing Notes:

  • Tested across 4 different LLM providers (OpenAI, Anthropic, Gemini, Moonshot)
  • Comprehensive test coverage added for new features
  • Manual testing performed for all UI changes
  • Database migrations tested on clean and existing databases

Known Limitations:

  • Analytics data can grow large over time - consider implementing data retention policies
  • PDF generation currently supports basic markdown - complex formatting may not render perfectly
  • Some Langfuse advanced features require Langfuse server version 2.0+

Migration Path:
This is a non-breaking update. Existing deployments can upgrade without data loss or configuration changes. New features are opt-in via environment variables.

Note

Medium Risk
Touches LLM provider configuration and reasoning/tool-call handling plus logging/tracing interfaces, which can cause provider-specific runtime errors or compilation breakages if any integration points are missed.

Overview
Adds custom LLM compatibility options via new env vars LLM_SERVER_PROVIDER (model-name prefixing for LiteLLM-style proxies) and LLM_SERVER_PRESERVE_REASONING (keep reasoning content across turns for providers like Moonshot), wiring them through .env.example, the installer wizard UI/model, and ctester (new -name flag).

Updates tool/log plumbing to carry taskID/subtaskID into screenshot + terminal log providers and tool constructors, and adjusts Langfuse tracing calls to newer API method names. Also refreshes packaging/dev ergonomics: bumps final Docker base image to alpine:3.23.3, tweaks included provider config files, updates VSCode launch/settings defaults, and adds/refreshes extensive docs (notably new backend/docs/analytics_api.md plus chain summarization/provider reasoning requirements).

Written by Cursor Bugbot for commit 6096ab1528. This will update automatically on new commits. Configure here.

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/vxcontrol/pentagi/pull/102 **Author:** [@asdek](https://github.com/asdek) **Created:** 2/18/2026 **Status:** ✅ Merged **Merged:** 2/18/2026 **Merged by:** [@asdek](https://github.com/asdek) **Base:** `master` ← **Head:** `feat/langchaingo_update` --- ### 📝 Commits (10+) - [`e6cbf4a`](https://github.com/vxcontrol/pentagi/commit/e6cbf4a2fd24d67beffa78e9b9cd89c4f8e5fcb1) feat: major langchaingo dependency update and add product working analysis API - [`8f1f9ee`](https://github.com/vxcontrol/pentagi/commit/8f1f9ee0dcd1a771b1d8f8c893bb05543c3259c9) feat: update graphiti-go-client dependency and enhance observability integration - [`fd60a8f`](https://github.com/vxcontrol/pentagi/commit/fd60a8f6ccd1490e90a0605b20b7d69de95c2cbd) feat: add analytics service and endpoints for usage statistics - [`e193390`](https://github.com/vxcontrol/pentagi/commit/e193390c0d95fde54f99d16b93cb8c31a8ee4a2b) fix: typo in graphiti source description - [`d307fa1`](https://github.com/vxcontrol/pentagi/commit/d307fa16e7cfccd6f0330718e8e80a455b38a29b) feat: enhance summarization handling for reasoning signatures - [`e68e237`](https://github.com/vxcontrol/pentagi/commit/e68e2374028a9638239912106694552ca1f92409) feat: enhance summarization functionality for multi-provider compatibility - [`efd966c`](https://github.com/vxcontrol/pentagi/commit/efd966c0635f645d7ababce0b1dc46b5bd5aded1) refactor: replace newID() with newSpanID() for W3C compliance - [`16bd43c`](https://github.com/vxcontrol/pentagi/commit/16bd43cb83bfb0445eddd1f6cf413768a85bf060) fix: chain summarization with critical guarantees - [`4065300`](https://github.com/vxcontrol/pentagi/commit/40653008961883c2994a5fdd4b33aaaa8aaf8bb1) chore: update base image in Dockerfile from alpine:3.22.1 to alpine:3.23.3 - [`d84d93b`](https://github.com/vxcontrol/pentagi/commit/d84d93bae15aa4c5bfd8372c5bd263389a79c66a) chore: update model and pricing in moonshot provider configuration ### 📊 Changes **277 files changed** (+99895 additions, -18833 deletions) <details> <summary>View changed files</summary> 📝 `.env.example` (+2 -0) 📝 `.vscode/launch.json` (+5 -3) 📝 `.vscode/settings.json` (+2 -2) 📝 `Dockerfile` (+8 -7) 📝 `README.md` (+104 -57) 📝 `backend/cmd/ctester/main.go` (+4 -0) 📝 `backend/cmd/ftester/mocks/logs.go` (+17 -1) 📝 `backend/cmd/ftester/worker/executor.go` (+6 -0) 📝 `backend/cmd/ftester/worker/tester.go` (+6 -5) 📝 `backend/cmd/installer/wizard/controller/controller.go` (+19 -3) 📝 `backend/cmd/installer/wizard/locale/locale.go` (+9 -0) 📝 `backend/cmd/installer/wizard/models/llm_provider_form.go` (+50 -0) ➕ `backend/docs/analytics_api.md` (+1005 -0) 📝 `backend/docs/chain_ast.md` (+106 -1) 📝 `backend/docs/chain_summary.md` (+208 -10) 📝 `backend/docs/config.md` (+158 -144) 📝 `backend/docs/controller.md` (+1 -1) 📝 `backend/docs/database.md` (+587 -47) 📝 `backend/docs/langfuse.md` (+534 -37) ➕ `backend/docs/llms_how_to.md` (+1865 -0) _...and 80 more files_ </details> ### 📄 Description <!-- Thank you for your contribution to PentAGI! Please fill out this template completely to help us review your changes effectively. Any PR that does not include enough information may be closed at maintainers' discretion. --> ### Description of the Change This PR introduces a major update to the langchaingo dependency and implements comprehensive observability improvements across the system. It also adds analytics capabilities, enhances chain summarization logic, and includes various UI/UX improvements. #### Problem The system required several critical improvements: 1. Outdated langchaingo dependency limiting provider compatibility and features 2. Inconsistent observability and tracing across components 3. Lack of system-wide analytics and usage statistics 4. Chain summarization issues with multi-provider support (Gemini, Anthropic, Kimi/Moonshot) 5. Missing W3C Trace Context compliance for distributed tracing 6. Various UI/UX issues affecting user experience #### Solution **Backend Infrastructure:** - Updated langchaingo to the latest version with enhanced provider support - Standardized observability implementation across all components with W3C-compliant trace/span IDs - Introduced comprehensive analytics service with endpoints for system-wide and flow-specific usage statistics - Enhanced chain summarization algorithm with critical guarantees for reasoning preservation across different LLM providers - Added support for `LLM_SERVER_PROVIDER` and `LLM_SERVER_PRESERVE_REASONING` environment variables - Updated Langfuse API client with new endpoints (annotation queues, blob storage integrations, LLM connections, etc.) - Improved tool call ID generation and validation with function name support **Frontend Improvements:** - Added PDF generation capabilities using @react-pdf/renderer for flow reports - Enhanced theme handling with proper system theme detection and persistence - Improved authentication flow with safe return URL handling to prevent open redirects - Better terminal component theme integration **DevOps & Configuration:** - Updated base Alpine image from 3.22.1 to 3.23.3 - Enhanced provider configurations (Moonshot model updates, additional provider files) - Improved VSCode debug and settings configurations **Database & Analytics:** - Added new analytics models and queries for comprehensive usage tracking - Implemented tracking fields for better observability (tool calls, token usage, costs) - Created migration scripts for new analytics tables <!-- Enter any applicable Issue number(s) here that will be closed/resolved by this PR. --> Closes # ### Type of Change <!-- Mark with an `x` all options that apply --> - [x] 🐛 Bug fix (non-breaking change which fixes an issue) - [x] 🚀 New feature (non-breaking change which adds functionality) - [ ] 💥 Breaking change (fix or feature that would cause existing functionality to not work as expected) - [x] 📚 Documentation update - [x] 🔧 Configuration change - [x] 🧪 Test update - [ ] 🛡️ Security update ### Areas Affected <!-- Mark with an `x` all components that are affected --> - [x] Core Services (Frontend UI/Backend API) - [x] AI Agents (Researcher/Developer/Executor) - [ ] Security Tools Integration - [x] Memory System (Vector Store/Knowledge Base) - [x] Monitoring Stack (Grafana/OpenTelemetry) - [x] Analytics Platform (Langfuse) - [x] External Integrations (LLM/Search APIs) - [x] Documentation - [x] Infrastructure/DevOps ### Testing and Verification The changes have been tested across multiple LLM providers and system components to ensure compatibility and stability. #### Test Configuration ```yaml PentAGI Version: dev (feat/langchaingo_update branch) Docker Version: 24.x+ Host OS: Linux/macOS LLM Provider: OpenAI, Anthropic, Google Gemini, Moonshot Enabled Features: [Langfuse, Grafana, Analytics] ``` #### Test Steps 1. Updated all Go dependencies and verified compilation 2. Ran comprehensive test suites for: - Chain summarization with different providers (Anthropic, Gemini, Moonshot) - Tool call ID generation and validation - Analytics service endpoints - Observability integration (Langfuse traces, spans, events) 3. Tested frontend changes: - PDF report generation from flow results - Theme switching (light/dark/system) - Authentication flow with return URL handling 4. Verified database migrations execute successfully 5. Tested LLM provider compatibility with new langchaingo version 6. Validated W3C Trace Context compliance in distributed tracing 7. Confirmed analytics endpoints return correct usage statistics #### Test Results - All existing unit tests pass - New test cases added for: - Chain summarization with reasoning signatures - Analytics data conversion - Tool call ID pattern generation - Langfuse observation conversion - Manual testing confirmed PDF generation works correctly - Theme handling properly respects system preferences - Authentication redirects work securely without open redirect vulnerabilities - Analytics API returns accurate system-wide and flow-specific metrics ### Security Considerations **Enhanced Security Measures:** - Implemented `getSafeReturnUrl` utility to prevent open redirect vulnerabilities in authentication flow - All return URLs are validated to ensure they are relative paths only - Updated Alpine base image to 3.23.3 to include latest security patches **Dependency Updates:** - Updated langchaingo dependency with latest security fixes - Updated Langfuse API client with improved error handling and security features - All new dependencies have been reviewed for known vulnerabilities **No Breaking Security Changes:** - Existing authentication and authorization mechanisms remain intact - Session handling unchanged - API token management unaffected by observability changes ### Performance Impact **Improvements:** - Enhanced chain summarization algorithm is now idempotent and more efficient - Memoized theme detection function reduces unnecessary re-renders in terminal component - Optimized observability event batching in Langfuse integration - Better memory management in tool call ID generation **New Analytics Overhead:** - Analytics service adds minimal overhead (<1% CPU) for usage tracking - Database queries optimized with proper indexing (existing tracking fields) - Analytics endpoints use efficient aggregation queries **Testing:** - Performance tested with large message chains (1000+ messages) - Verified summarization maintains sub-second latency for typical flows - Analytics queries tested with 10K+ flow records - No measurable impact on LLM request/response latency **Recommendations:** - Analytics data should be periodically archived for long-running systems - Consider adding database indices on `created_at` fields for faster analytics queries at scale ### Documentation Updates <!-- Note any documentation changes required by this PR --> - [x] README.md updates (new LLM_SERVER_PROVIDER and LLM_SERVER_PRESERVE_REASONING variables) - [x] API documentation updates (Swagger/OpenAPI docs for analytics endpoints) - [x] Configuration documentation updates (backend/docs/config.md) - [x] GraphQL schema updates (analytics-related queries and types) - [x] Other: - Added `backend/docs/analytics_api.md` - comprehensive analytics API documentation - Added `backend/docs/llms_how_to.md` - LLM provider configuration guide - Updated `backend/docs/observability.md` - enhanced observability documentation - Updated `backend/docs/chain_summary.md` - chain summarization algorithm details - Updated `backend/docs/chain_ast.md` - AST handling for reasoning signatures - Updated `.env.example` with new configuration options ### Deployment Notes **New Environment Variables:** ```bash # Optional: Specify server-side LLM provider for specific features LLM_SERVER_PROVIDER=openai # or anthropic, gemini, etc. # Optional: Preserve reasoning content for providers that support it (Kimi/Moonshot) LLM_SERVER_PRESERVE_REASONING=true ``` **Database Migrations:** The following migrations will run automatically on deployment: - `20260128_153000_tool_call_id_template.sql` - Adds tool call ID template tracking - `20260129_120000_add_tracking_fields.sql` - Adds analytics tracking fields **Docker Image Updates:** - Base Alpine image updated to 3.23.3 - Additional provider configuration files included in the image **Recommended Deployment Steps:** 1. Pull latest changes from `feat/langchaingo_update` branch 2. Update `.env` file with new optional variables (if needed) 3. Run `docker-compose pull` to get updated base images 4. Run `docker-compose down && docker-compose up -d` for clean restart 5. Verify database migrations completed successfully (check logs) 6. Test analytics endpoints: `GET /api/v1/analytics/usage` 7. Verify Langfuse integration still working (if enabled) **Backward Compatibility:** - All changes are backward compatible - Existing flows and data remain unaffected - New environment variables are optional with sensible defaults - GraphQL schema changes are additive only **Rollback Plan:** If issues occur, rollback is straightforward: 1. Checkout previous commit 2. Rebuild containers: `docker-compose up -d --build` 3. Database migrations are forward-compatible (no rollback needed) ### Checklist <!--- Go over all the following points, and put an `x` in all the boxes that apply. --> #### Code Quality - [x] My code follows the project's coding standards - [x] I have added/updated necessary documentation - [x] I have added tests to cover my changes - [x] All new and existing tests pass - [x] I have run `go fmt` and `go vet` (for Go code) - [x] I have run `npm run lint` (for TypeScript/JavaScript code) #### Security - [x] I have considered security implications - [x] Changes maintain or improve the security model - [x] Sensitive information has been properly handled #### Compatibility - [x] Changes are backward compatible - [x] Breaking changes are clearly marked and documented - [x] Dependencies are properly updated #### Documentation - [x] Documentation is clear and complete - [x] Comments are added for non-obvious code - [x] API changes are documented ### Additional Notes **Key Changes Summary:** 1. **Langchaingo Update (16 commits worth of work):** - Major dependency update bringing compatibility with latest LLM providers - Enhanced provider support for Anthropic, OpenAI, Gemini, Moonshot/Kimi - Improved reasoning signature handling across different providers 2. **Observability Enhancements:** - W3C Trace Context compliance for distributed tracing - Standardized span/trace ID handling throughout the codebase - Enhanced Langfuse integration with new observation types (agent, chain, tool, retriever, embedding, evaluator, guardrail) - Added 30+ new Langfuse API client methods 3. **Analytics Infrastructure:** - New comprehensive analytics service with system-wide and per-flow statistics - Tracking for token usage, costs, tool calls, and execution times - RESTful API endpoints for analytics data retrieval 4. **Chain Summarization Improvements:** - Critical guarantees for preserving last N QA sections - Idempotent operation - multiple summarizations don't alter already summarized content - Provider-specific handling for reasoning signatures (Gemini, Anthropic, Kimi) - Enhanced testing coverage for edge cases 5. **Frontend Enhancements:** - PDF export functionality for flow reports using @react-pdf/renderer - Improved theme system with proper system theme detection - Secure authentication flow preventing open redirects **Review Focus Areas:** - Chain summarization logic (complex but critical for multi-provider support) - Analytics service database queries and performance - Observability integration changes (verify Langfuse still works as expected) - PDF generation quality and formatting - Security of authentication return URL handling **Testing Notes:** - Tested across 4 different LLM providers (OpenAI, Anthropic, Gemini, Moonshot) - Comprehensive test coverage added for new features - Manual testing performed for all UI changes - Database migrations tested on clean and existing databases **Known Limitations:** - Analytics data can grow large over time - consider implementing data retention policies - PDF generation currently supports basic markdown - complex formatting may not render perfectly - Some Langfuse advanced features require Langfuse server version 2.0+ **Migration Path:** This is a non-breaking update. Existing deployments can upgrade without data loss or configuration changes. New features are opt-in via environment variables. <!-- CURSOR_SUMMARY --> --- > [!NOTE] > **Medium Risk** > Touches LLM provider configuration and reasoning/tool-call handling plus logging/tracing interfaces, which can cause provider-specific runtime errors or compilation breakages if any integration points are missed. > > **Overview** > Adds **custom LLM compatibility options** via new env vars `LLM_SERVER_PROVIDER` (model-name prefixing for LiteLLM-style proxies) and `LLM_SERVER_PRESERVE_REASONING` (keep reasoning content across turns for providers like Moonshot), wiring them through `.env.example`, the installer wizard UI/model, and `ctester` (new `-name` flag). > > Updates tool/log plumbing to carry `taskID`/`subtaskID` into screenshot + terminal log providers and tool constructors, and adjusts Langfuse tracing calls to newer API method names. Also refreshes packaging/dev ergonomics: bumps final Docker base image to `alpine:3.23.3`, tweaks included provider config files, updates VSCode launch/settings defaults, and adds/refreshes extensive docs (notably new `backend/docs/analytics_api.md` plus chain summarization/provider reasoning requirements). > > <sup>Written by [Cursor Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit 6096ab15289e60252d1b1a40fa9d45e17e17e377. This will update automatically on new commits. Configure [here](https://cursor.com/dashboard?tab=bugbot).</sup> <!-- /CURSOR_SUMMARY --> --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
yindo added the pull-request label 2026-06-06 22:09:28 -04:00
yindo closed this issue 2026-06-06 22:09:28 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
enhancement
help wanted
pull-request
Mirrored from GitHub Pull Request
 question
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: vxcontrol/pentagi#161
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?