Skip to content

Latest commit

 

History

History
124 lines (96 loc) · 5.53 KB

CHANGELOG.md

File metadata and controls

124 lines (96 loc) · 5.53 KB

Changelog

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

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

1.2.0 - 2026-02-14

Added

  • 4 New Skills

    • database-design - Schema design, indexing strategies, query optimization, migrations
    • devops-infrastructure - Docker, CI/CD, deployment strategies, IaC, monitoring
    • error-handling - Error patterns by language, structured logging, retry/circuit breakers
    • security-patterns - JWT/OAuth auth, RBAC, secrets management, CORS, rate limiting

  • 7 New Commands

    • /project-starter:refactor-guided - 4-phase systematic refactoring with safety rules
    • /project-starter:dependency-upgrade - Safe dependency upgrades with rollback
    • /project-starter:plan - Manus-style persistent PLAN.md with phase tracking
    • /project-starter:tutorial - Interactive guided tutorial for first-time users
    • /project-starter:bootstrap-repo - 10-agent parallel repo exploration and CODEBASE.md generation
    • /project-starter:save-session-learnings - Persist session discoveries to CLAUDE.md/AGENTS.md
    • /project-starter:metrics - View agent performance metrics and session history

  • 5 New Hooks

    • pre-commit-check.py - Detects debug statements, temp markers, large file content
    • branch-protection.sh - Warns on git operations targeting protected branches
    • typescript-check.py - Runs tsc --noEmit after editing .ts/.tsx files
    • suggest-doc-updates.py - Suggests CLAUDE.md updates when significant changes detected
    • track-metrics.py - Logs session telemetry to .claude/agent-metrics.jsonl

  • On-Demand Context Loading

    • All 14 skills now include "When to Load" sections with trigger/skip conditions

  • CI/CD Pipeline

    • .github/workflows/validate.yml - Plugin validation (JSON, scripts, frontmatter, links)

Changed

  • All 7 Agents Upgraded with:
    • Action-first directive (act before explaining)
    • Adaptive effort scaling (Instant/Light/Deep/Exhaustive)
    • Adversarial self-review protocol (5-point checklist)
    • Intellectual honesty framework (Certain/Likely/Uncertain)
    • Expanded auto-trigger keywords
    • Paired WRONG/CORRECT anti-pattern examples for each domain
  • web-design-guidelines skill rewritten as self-contained 190-line reference (was 37-line external URL dependency)
  • Hook error messages now include actionable remediation suggestions across all hooks
  • debugger agent gains escalation protocol (after 3 failed attempts: web search, backtrack)
  • security-auditor agent gains dependency vulnerability checking section

1.1.0 - 2025-01-07

Added

  • Parallel Execution Support

    • Orchestrator can now spawn N subagents simultaneously for N independent tasks
    • Uses run_in_background: true with Task tool for concurrent execution
    • Results collected via TaskOutput after all agents complete
    • Roughly Nx faster than sequential execution

  • New Commands

    • /project-starter:parallel-review - Review multiple directories in parallel
    • /project-starter:parallel-analyze - Analyze code from architecture, security, performance, and testing perspectives simultaneously

  • New Skill

    • parallel-execution - Patterns for spawning dynamic subagents, best practices for parallelization, and TodoWrite integration

Changed

  • Updated orchestrator agent with Parallel Execution Protocol section
  • Enhanced verify-changes command with explicit parallel syntax
  • Added parallel execution examples and diagrams to documentation

1.0.0 - 2025-01-01

Added

  • 7 Specialized Agents

    • orchestrator - Coordinate complex multi-step tasks
    • code-reviewer - Review code quality and best practices
    • debugger - Systematic bug investigation and fixing
    • docs-writer - Create technical documentation
    • security-auditor - Security vulnerability detection
    • refactorer - Code structure improvements
    • test-architect - Design comprehensive test strategies

  • 6 Knowledge Skills

    • project-analysis - Understand any codebase structure and patterns
    • testing-strategy - Design test approaches (unit, integration, E2E)
    • architecture-patterns - System design guidance
    • performance-optimization - Speed up applications, identify bottlenecks
    • git-workflow - Version control best practices
    • api-design - REST/GraphQL API patterns

  • 4 Output Styles

    • /project-starter:architect - System design mode
    • /project-starter:rapid - Fast development mode
    • /project-starter:mentor - Learning/teaching mode
    • /project-starter:review - Code review mode

  • 8 Automation Hooks

    • Security scan (blocks commits with potential secrets)
    • File protection (blocks edits to lock files, .env, .git)
    • Auto-format (runs prettier/black/gofmt based on file type)
    • Command logging (logs to .claude/command-history.log)
    • Environment check (validates Node.js, Python, Git)
    • Prompt analysis (suggests appropriate agents)
    • Input notification (desktop notification when input needed)
    • Complete notification (desktop notification when task finishes)

  • Cross-platform notification support (macOS, Linux, Windows)

  • Comprehensive documentation (README, PERMISSIONS, MCP servers guide)

  • MIT License