Release v3.5.10

Version bump for release

Release v3.5.9

[3.5.9] - 2026-01-16

Changed

• Version bump for release

Release v3.5.8

Bump version to 3.5.8

Release v3.5.7

Changelog

All notable changes to TappsCodingAgents will be documented in this file.

The format is based on Keep a Changelog,
and this project adheres to Semantic Versioning.

[Unreleased]

[3.5.7] - 2026-01-16

Added

• Full SDLC Workflow Investigation Documentation - Comprehensive investigation results

  • Added docs/FULL_SDLC_WORKFLOW_ISSUE_INVESTIGATION.md - Investigation framework for Full SDLC workflow issues
  • Added docs/FULL_SDLC_INVESTIGATION_RESULTS.md - Investigation findings confirming Full SDLC workflow is working correctly
  • Clarified use case differences between *full and *build workflows

• Workflow File Structure Documentation - Documentation for workflow file structure requirements

  • Added docs/WORKFLOW_FILE_STRUCTURE_ANALYSIS.md - Analysis of Full SDLC workflow file structure expectations
  • Added docs/SIMPLE_MODE_BUILD_QUICK_START.md - Quick start guide for Simple Mode *build workflow
  • Added docs/RECOMMENDATIONS_EXECUTION_GUIDE.md - Practical guide for executing workflow recommendations

Fixed

• Continuous Bug Fix - Fixed error handling and pytest collection error

  • Improved error handling in continuous bug fix workflow
  • Fixed pytest collection error handling

• Background Agents Configuration - Optimized background agents configuration

  • Improved configuration based on recommendations
  • Better performance and reliability

• Documentation Accuracy - Updated documentation for accuracy

  • Updated Cursor Skills count to 15 (14 agent skills + simple-mode)
  • Verified init and init --reset command documentation accuracy

[3.5.6] - 2026-01-16

Added

• Continuous Bug Finder and Fixer - New feature for automated bug detection and fixing
  • Continuously runs tests, detects bugs, fixes them using bug-fix-agent, and commits fixes automatically
  • Stops when no bugs are found or max iterations reached
  • Supports one-per-bug and batch commit strategies
  • Configurable max iterations and test paths

Fixed

• GitHub Actions CI - Fixed coverage threshold and summary script

  • Adjusted coverage threshold to match project requirements
  • Fixed summary script for better CI reporting

• Enhancer Analysis - Fixed analysis stage to reliably populate all fields

  • Added fallback mechanism to ensure all analysis fields are populated
  • Prevents "unknown" values when analyst/LLM is unavailable
  • Improved reliability of prompt enhancement workflow

• Bug Fixing Workflow - Fixed debugger command and git worktree cleanup

  • Corrected debugger command invocation in bug fixing workflow
  • Improved git worktree cleanup to prevent resource leaks

[3.5.5] - 2026-01-16

Fixed

• Framework Change Detector - Fixed 5 bugs in framework_change_detector.py
  • Fixed incorrect modified_agents logic that included all existing agents instead of only modified ones
  • Added exception handling for PermissionError and OSError when scanning agent directories (Windows compatibility)
  • Fixed regex injection vulnerability by escaping special characters in agent names
  • Improved path construction robustness with directory structure validation
  • Fixed command extraction to preserve order using dict.fromkeys() instead of set()

Tests

• Framework Change Detector Tests - Added comprehensive unit test coverage for all 5 bug fixes
  • Added test for modified_agents bug fix
  • Added tests for PermissionError/OSError handling
  • Added test for regex escaping with special characters
  • Added test for improved path construction
  • Added test for command order preservation
  • All 18 tests passing (13 existing + 5 new)

[3.5.4] - 2026-01-11

Added

• JSON Schema Generation - Complete JSON Schema files for all artifact models

  • Generated 29 JSON Schema files in schemas/1.0/ directory
  • Schema generation script: scripts/generate_artifact_schemas.py
  • Schema index file for easy reference
  • Schema documentation in schemas/README.md
  • Supports JSON Schema validation for artifact data

• Pydantic Migration Documentation - Comprehensive migration documentation

  • Complete migration guide: docs/IMPLEMENTATION/PYDANTIC_MIGRATION_GUIDE.md
  • Research documentation: docs/IMPLEMENTATION/pydantic_migration_research.md
  • Migration status tracking: docs/IMPLEMENTATION/pydantic_migration_status.md
  • Backward compatibility analysis documentation

Changed

• API Documentation - Enhanced API documentation with artifact model details
  • Added "Data Models" section to docs/API.md
  • Documented all Pydantic artifact models
  • Documented enum types and structured metadata models
  • Added JSON Schema usage examples
  • Added migration guide references

Documentation

• Complete Pydantic Migration - All artifact models migrated to Pydantic BaseModel
  • All 10 artifact types now use Pydantic v2 BaseModel
  • Type-safe enums (Priority, ArtifactStatus, RiskLevel, OperationType, StoryStatus)
  • Structured metadata models (PlanDetails, TaskInputs, TaskResults, RetryPolicy)
  • Unified Story model combining UserStory and Epic.Story features
  • Epic models migrated to Pydantic
  • Full backward compatibility maintained via from_dict() methods
  • All tests passing without modification

[3.5.3] - 2026-01-09

Fixed

• Framework Change Detector - Improved handling when agents directory not found

  • Changed warning to debug level when package is installed (expected behavior)
  • Added clarifying message that this is normal when using installed package
  • Reduces noise in logs for users running from installed packages

• Error Handling - Improved error message extraction for missing agent results

  • Better detection of missing agent execution results
  • More actionable error messages indicating when agent results are not returned
  • Improved diagnostics for debugging workflow execution failures

[3.5.2] - 2026-01-09

Changed

• Artifact System Migration - Migrated artifacts from dataclasses to Pydantic BaseModel
  • Improved type safety and validation for workflow artifacts
  • Better integration with Pydantic v2 patterns
  • Enhanced serialization and deserialization capabilities

Added

• Progress Tracking Enhancement - Phase 3 implementation with Progress.txt and story-level granularity
  • Enhanced workflow progress tracking at story level
  • Improved visibility into workflow execution progress

Fixed

• Code Quality - Fixed duplicate import patterns across codebase
  • Removed redundant imports to improve code clarity
  • Enhanced code maintainability

Documentation

• Documentation Updates - Updated command-reference.mdc and README.md
  • Improved command reference documentation
  • Enhanced README with latest information

[3.5.1] - 2026-01-09

Fixed

• Missing Evaluator Agent - Added evaluator to AGENT_TYPES list and CLI help text
  • Evaluator agent was missing from skill-template command agent type choices
  • Added evaluator to CLI help text in AGENT COMMANDS section
  • Updated skill-template help text to include evaluator in agent type list

Changed

• Init Process Documentation - Comprehensive review and optimization of all init-related documentation
  • Fixed missing evaluator skill in FRAMEWORK_SKILLS list (init_project.py)
  • Added missing simple-mode workflow presets to FRAMEWORK_WORKFLOW_PRESETS
  • Updated all documentation files to reflect 14 agents + simple-mode (15 total)
  • Added evaluator agent documentation to command-reference.mdc and agent-capabilities.mdc
  • Updated project-context.mdc with accurate agent count and Cursor Skills reference
  • Added workflow execution flags documentation to workflow-presets.mdc
  • Ensured consistency between resource files and installed .cursor/rules/ files
  • All framework file counts now match actual resources (15 skills, 7 rules, 8 presets)

[3.5.0] - 2026-01-09

Added

• TypeScript Security Analysis - Real security pattern detection for TypeScript/JavaScript files

  • Detects dangerous patterns: eval(), innerHTML, dangerouslySetInnerHTML, document.write, etc.
  • CWE IDs included for all security patterns (CWE-94, CWE-79)
  • SecurityIssue dataclass for structured security reporting
  • Security score now reflects actual issues found (not default 5.0)

• Improver Auto-Apply Option - Automatically apply code improvements

  • New --auto-apply flag: Apply improvements with automatic backup
  • New --preview flag: Show diff without modifying file
  • Timestamped backups stored in .tapps-agents/backups/
  • Verification review runs after auto-apply to confirm improvements
  • DiffResult dataclass with unified diff and line statistics

• Score Explanation Mode - Detailed explanations for all scores

  • New --explain flag for reviewer score command
  • Explanations include: reason, identified issues, recommendations
  • Tool availability status (ESLint, TypeScript compiler)
  • Works for both Python and TypeScript/JavaScript files

• Traceability Documentation - Requirements to implementation mapping

  • TRACEABILITY_MATRIX.md - Links requirements → stories → implementation → tests
  • STORY_VERIFICATION_CHECKLIST.md - Acceptance criteria verification
  • Gherkin-to-test mapping in test files

Tests

• 56 new tests for TypeScript enhancement features
• tests/agents/reviewer/test_typescript_security.py (26 tests)
• tests/agents/improver/test_auto_apply.py (18 tests)
• Acceptance criteria tests mapped to Gherkin scenarios

Documentation

• docs/TYPESCRIPT_SUPPORT.md - Complete TypeScript support guide
• docs/RECOMMENDATIONS_IMPLEMENTATION_SUMMARY.md - Implementation report
• Updated workflow documentation in docs/workflows/simple-mode/

[3.4.1] - 2026-01-09

Fixed

• **Context7 API ...

Release v3.5.6

[3.5.6] - 2026-01-16

Added

• Continuous Bug Finder and Fixer - New feature for automated bug detection and fixing
  • Continuously runs tests, detects bugs, fixes them using bug-fix-agent, and commits fixes automatically
  • Stops when no bugs are found or max iterations reached
  • Supports one-per-bug and �atch commit strategies
  • Configurable max iterations and test paths

Fixed

• GitHub Actions CI - Fixed coverage threshold and summary script

  • Adjusted coverage threshold to match project requirements
  • Fixed summary script for better CI reporting

• Enhancer Analysis - Fixed analysis stage to reliably populate all fields

  • Added fallback mechanism to ensure all analysis fields are populated
  • Prevents unknown values when analyst/LLM is unavailable
  • Improved reliability of prompt enhancement workflow

• Bug Fixing Workflow - Fixed debugger command and git worktree cleanup

  • Corrected debugger command invocation in bug fixing workflow
  • Improved git worktree cleanup to prevent resource leaks

Release v3.5.5

Fixed

• Framework Change Detector - Fixed 5 bugs in \ramework_change_detector.py\
  • Fixed incorrect \modified_agents\ logic that included all existing agents instead of only modified ones
  • Added exception handling for \PermissionError\ and \OSError\ when scanning agent directories (Windows compatibility)
  • Fixed regex injection vulnerability by escaping special characters in agent names
  • Improved path construction robustness with directory structure validation
  • Fixed command extraction to preserve order using \dict.fromkeys()\ instead of \set()\

Tests

• Framework Change Detector Tests - Added comprehensive unit test coverage for all 5 bug fixes
  • Added test for \modified_agents\ bug fix
  • Added tests for \PermissionError/\OSError\ handling
  • Added test for regex escaping with special characters
  • Added test for improved path construction
  • Added test for command order preservation
  • All 18 tests passing (13 existing + 5 new)

Release v3.5.3

Fixed

• Framework Change Detector - Improved handling when agents directory not found

  • Changed warning to debug level when package is installed (expected behavior)
  • Added clarifying message that this is normal when using installed package
  • Reduces noise in logs for users running from installed packages

• Error Handling - Improved error message extraction for missing agent results

  • Better detection of missing agent execution results
  • More actionable error messages indicating when agent results are not returned
  • Improved diagnostics for debugging workflow execution failures

v3.5.1 - Documentation fixes and evaluator agent bug fix

Fixed

• Missing Evaluator Agent - Added evaluator to AGENT_TYPES list and CLI help text
  • Evaluator agent was missing from skill-template command agent type choices
  • Added evaluator to CLI help text in AGENT COMMANDS section
  • Updated skill-template help text to include evaluator in agent type list

Changed

• Init Process Documentation - Comprehensive review and optimization of all init-related documentation
  • Fixed missing evaluator skill in FRAMEWORK_SKILLS list (init_project.py)
  • Added missing simple-mode workflow presets to FRAMEWORK_WORKFLOW_PRESETS
  • Updated all documentation files to reflect 14 agents + simple-mode (15 total)
  • Added evaluator agent documentation to command-reference.mdc and agent-capabilities.mdc
  • Updated project-context.mdc with accurate agent count and Cursor Skills reference
  • Added workflow execution flags documentation to workflow-presets.mdc
  • Ensured consistency between resource files and installed .cursor/rules/ files
  • All framework file counts now match actual resources (15 skills, 7 rules, 8 presets)

v3.5.0 - TypeScript Enhancement Suite

🚀 New Features

TypeScript Security Analysis

• Real security pattern detection for TypeScript/JavaScript files
• Detects dangerous patterns: eval(), innerHTML, dangerouslySetInnerHTML, document.write, etc.
• CWE IDs included for all security patterns (CWE-94, CWE-79)
• SecurityIssue dataclass for structured security reporting
• Security score now reflects actual issues found

Improver Auto-Apply Option

• New --auto-apply flag: Apply improvements with automatic backup
• New --preview flag: Show diff without modifying file
• Timestamped backups stored in .tapps-agents/backups/
• Verification review runs after auto-apply
• DiffResult dataclass with unified diff and line statistics

Score Explanation Mode

• New --explain flag for reviewer score command
• Explanations include: reason, identified issues, recommendations
• Tool availability status (ESLint, TypeScript compiler)
• Works for both Python and TypeScript/JavaScript files

📚 Documentation

• docs/TYPESCRIPT_SUPPORT.md - Complete TypeScript support guide
• docs/TRACEABILITY_MATRIX.md - Requirements to implementation mapping
• docs/STORY_VERIFICATION_CHECKLIST.md - Acceptance criteria verification

✅ Tests

• 56 new tests for TypeScript enhancement features
• tests/agents/reviewer/test_typescript_security.py (26 tests)
• tests/agents/improver/test_auto_apply.py (18 tests)
• Acceptance criteria tests mapped to Gherkin scenarios

📖 Usage

# Score TypeScript with explanations
tapps-agents reviewer score src/app.tsx --explain

# Auto-apply improvements with backup
tapps-agents improver improve-quality src/app.tsx --auto-apply

# Preview changes without modifying
tapps-agents improver improve-quality src/app.tsx --preview

📦 Installation

pip install tapps-agents==3.5.0

Release v3.4.1

Fixed

• Context7 API Authentication - Corrected authentication header in HTTP fallback client
  • Changed from X-API-Key header to Authorization: Bearer token
  • Verified Bearer token correctly authenticates (Quota Tier: free/pro vs anonymous)
  • API calls now properly authenticate and respect quota limits
  • KB-first cache lookup continues to work correctly for cached libraries

Added

• Context7 API Key Management Utility - New scripts/update_context7_key.py script
  • Store, verify, and test Context7 API keys
  • Loads keys from encrypted storage automatically
  • Tests API connection with proper authentication

Documentation

• Updated docs/CONTEXT7_QUOTA_FIX.md with correct authentication method
• Documented verified authentication patterns and quota handling