Releases: thoeltig/claude-code-toolkit
Releases · thoeltig/claude-code-toolkit
v1.11.0.0
Added
- project-intel plugin (v1.0.0.0): Lightweight reconnaissance system that provides direction before exploration
/scancommand: Generate semantic summaries of project files and directories (one-time setup)/querycommand: Search summaries by semantic relevance before reading files- Semantic scoring system matching on purpose, summary, exports/imports, technologies, and path
- Wave-based parallel processing: Max 10 concurrent agents for large projects
- Persistent storage:
.knowledge/summaries.jsonsurvives across sessions - Directory and file summary generation with purpose, role, exports, imports, and technologies tracking
- File modification timestamp tracking for staleness detection
- Scope filtering: Limit queries to specific directories with
--scopeparameter - Result limiting: Control query results with
--maxparameter - Incremental updates: Re-scan merges new summaries with existing knowledge
- Best practices guide for maintenance strategy and query approaches
- Performance characteristics: Cost estimates for projects of different sizes
- Troubleshooting guide with common issues and solutions
- Comparison to alternatives: vs. Explore agent, Grep/Glob, direct file reading
- Design decision documentation: Why persistent storage, wave-based processing, semantic scoring, query-first workflow, and user-managed staleness
- Support for semantic matching on conceptual terms, not just keywords
- Project structure guidance: When to use project-intel, when to skip it, breakeven analysis
Full Changelog: v1.10.0.0...v1.11.0.0
v1.10.0.0
Added
- documentation plugin (v1.0.0.0): New plugin for creating and maintaining high-quality project documentation
- managing-documentation skill with 4 core workflows (WF1-WF4) for documentation lifecycle management
- WF1: Creating New Documentation - define purpose, plan structure, write for global audience, ensure inclusivity, validate quality
- WF2: Updating Existing Documentation - assess state, prioritize improvements, update content, verify changes
- WF3: Validating Documentation Quality - check content quality, clarity, inclusivity, completeness
- WF4: Applying Agile/Lean Principles - document late, update constantly, with purpose, choose best medium
- Support for multiple documentation types: guides, API docs, README files, architecture documentation
- ARID principles: Accept Repetition, Skimmable, Exemplary, Consistent, Current
- Global audience support: simple language, defined abbreviations, concrete examples, no idioms
- Inclusive language standards: no ableist, gendered, violent, or culturally specific language
- Quality checklists by document type (Guides, API Docs, README, Architecture)
- Validation patterns for problematic terms: ableist, gendered, violent language, time-based references, excessive claims, overused politeness
- Writing style principles: tone, voice, techniques, things to avoid
- Tool usage patterns: finding docs, validating content, understanding context, large-scale operations
- Common issues and fixes table with diagnosis and solutions
- Implementation approach with 6-step methodology
- Anti-patterns guide including outdated information, too much detail, unclear jargon, inconsistent terminology, cultural insensitivity
- Comprehensive documentation standards for clarity, accessibility, and inclusivity
- Support for Read, Write, Edit, Glob, Grep and Task tools
Changed
- fetch-full-content plugin (v1.1.0.0): Enhanced configuration flexibility
- Made folder argument optional with default cache directory
- Restricted allowed-tools to Python only (removed other tool types)
Full Changelog: v1.9.0.0...v1.10.0.0
v1.9.0.0
Added
- cross-platform-notification plugin (v1.0.0.0): New plugin for native system notifications
- Support for Windows, macOS, and Linux with platform-specific notification systems
- Windows toast notifications via PowerShell
- macOS notifications via terminal-notifier
- Linux notifications via notify-send
- Automatic fallback to console output when native notifications unavailable
- Hooks into Claude Code events for automated notifications
Changed
-
session-protocol plugin (v1.1.0.0):
- Breaking: Removed notification hook and related notification infrastructure
- Notification functionality separated into dedicated cross-platform-notification plugin for cleaner feature separation
- Simplified to focus on session continuity management
- Removed PreCompact hook
- SessionStart hook remains for auto-detection of existing session protocols
-
claude-code-capabilities plugin (v1.8.0.0):
- Removed
/list-skillsslash command - Claude Code v2.1.0 added a built-in/skillscommand, making the custom command obsolete
- Removed
Full Changelog: v1.8.1.0...v1.9.0.0
v1.8.0.0
Added
- fetch-full-content plugin (v1.0.0.0): Added
/fetch-full-contentcommand for complete web content retrieval- 100% content retrieval vs built-in WebFetch's 30-80% summarized content
- Filesystem caching to avoid redundant fetches on repeated analysis
- HTML to markdown conversion
- Support for multiple URLs and batch processing via URL files
⚠️ Security: NO prompt injection detection - only use on trusted sources (official docs, controlled content)- Recommendation: Use built-in WebFetch for untrusted sources
Full Changelog: v1.7.0.0...v1.8.0.0
v1.7.0.0
Added
- claude-code-capabilities plugin (v1.7.0): Added
/list-skillscommand for comprehensive skill enumeration- Query and display all available skills from personal (~/.claude/skills/), project (.claude/skills/), and plugin directories
- 5 output formats:
names(token-efficient),paths(file locations),list(human-readable, default),table(TSV for parsing),json(structured) - Powered by Python script (scripts/list-skills.py) for comprehensive skill discovery
- Usage:
/list-skills [--format names|paths|list|table|json]
Fixed
- Fixed duplicate marketplace.json files - consolidated to single root-level marketplace.json with correct field structure, enabling repository to function as GitHub marketplace source for Claude Code plugins
Full Changelog: v1.6.0.0...v1.7.0.0
v1.6.0.0 - Claude Code capabilities plugin extended
Added
- claude-code-capabilities plugin (v1.6.0): Added managing-mcps skill for comprehensive Model Context Protocol management
- 7 core workflows: Evaluating if logic should be MCP, creating servers with HTTP priority, analyzing MCPs, updating outdated MCPs, configuring connections, integrating via Messages API, using MCPs in conversations
- 13 supporting files with progressive disclosure:
- Core:
architecture-overview.md,creation-guide.md,analysis-framework.md,configuration-guide.md,security-best-practices.md - Advanced:
mcp-connector-api-integration.md(complete Messages API workflow),enterprise-mcp-configuration.md,environment-variable-expansion.md,plugin-mcp-servers.md,mcp-resources-and-prompts.md,oauth-authentication-flow.md,deprecation-notes.md - Reference:
examples.md(AWS Bedrock patterns, FastMCP templates)
- Core:
- Key features:
- HTTP transport priority: Explicit guidance - HTTP recommended for remote servers (better cloud support, more reliable)
- Messages API integration: Complete programmatic workflow with Python examples, MCPToolset configuration patterns (allowlist, denylist, mixed)
- AWS Bedrock patterns: Production-ready modular architecture using FastMCP framework with decorator-based tool registration
- FastMCP quick start: Minimal template for rapid Python MCP development
- Progressive disclosure: Core workflows in SKILL.md (557 lines), advanced topics loaded on-demand for token efficiency
- Cross-skill references: managing-plugins (plugin bundling), managing-prompts (prompt templates)
- Complete CLI coverage: All commands documented (add, add-json, add-from-claude-desktop, serve, reset-project-choices)
Full Changelog: v1.5.0.0...v1.6.0.0
v1.5.0.0 - Claude Code capabilities plugin extended
Added
-
claude-code-capabilities plugin (v1.5.0): Added managing-subagents skill for comprehensive subagent management
- 9 core workflows including resumable subagents (WF9)
- 10 supporting files with progressive disclosure
- Key features: 0-10 point delegation scoring, resumable agents with agentId tracking, 3 reference implementations, all 6 configuration fields with inline context, 5 permission modes, cross-skill references (managing-prompts, managing-plugins, managing-agent-skills)
- Management patterns: /agents interface, CLI configuration via --agents flag, plugin agent integration
- Complete alignment with official Claude Code subagent documentation (95%+ coverage)
- Operational focus: asks user for documentation when information unclear, no self-updating references
-
claude-code-capabilities plugin (v1.4.0): Added managing-slash-commands skill for custom slash command management
- 4 core workflows: creating, analyzing, suggesting conversion, updating commands
- 8 supporting files including SlashCommand tool reference, argument patterns, and examples
- Extended thinking support for complex reasoning tasks
- Plugin command and MCP command integration guidance
- Cross-skill references to managing-plugins, managing-mcps, and managing-prompts
Changed
- changelog plugin (v1.1.0):
- Removed HISTORY.md migration workflow (WF4) - simplified skill to focus on core changelog operations
Full Changelog: v1.4.0.0...v1.5.0.0
v1.4.0.0 - Claude Code capabilities plugin extended
-
Added claude-code-capabilities plugin (v1.3.0): Added managing-prompts skill prompt engineering mastery
- Complete skill for analyzing, creating, optimizing, and improving prompts using Claude 4.5 best practices
- 15 supporting files (14 + SKILL.md) covering all prompt engineering aspects
- 100% coverage of all 20 Anthropic prompt engineering documentation files
- Major improvements:
- Comprehensive 1-hour cache TTL documentation with pricing comparison, mixing rules, and decision flows
- Thinking blocks + prompt caching interaction fully documented (automatic caching, token counting, invalidation patterns)
- Expanded Structured Outputs API implementation with JSON schema requirements, comparison tables, and error handling
- Added crop tool reference for vision optimization
- Standardized terminology capitalization throughout
- 5 core workflows Analyzing prompts, creating prompts, optimizing prompts, updating outdated prompts, deciding logic extraction
- Advanced techniques: Extended thinking with budget management, prompt chaining architecture, consistency enforcement (Structured Outputs vs prefilling), guardrails implementation, context optimization with caching strategies
-
Added claude-code-capabilities plugin (v1.2.0): Added managing-plugins skill for creating, packing, bundling, and managing Claude Code plugins
- 5 core operations: Create plugin structure, bundle components by prefix, create marketplace config, validate plugins, pack for distribution
- Complete marketplace schema documentation including optional fields (strict, category, tags) and source object formats (GitHub, Git repositories)
- Team configuration workflow with
.claude/settings.jsonguidance for automatic plugin installation - Team testing workflow with 5-step validation process before rollout
- MCP server configuration fields documented (command, args, env, cwd)
- All 10 hook events documented with PermissionRequest added
- CLI commands reference: plugin validation, management, and marketplace operations
- Subagent terminology standardized and aligned with managing-agents skill
- Progressive disclosure: 5 supporting files with detailed specifications and real-world patterns
- 223-line SKILL.md with structured operations and actionable workflows
Full Changelog: v1.3.0.0...v1.4.0.0
v1.3.0.0 - Claude Code capabilities plugin extended
Added claude-code-capabilities plugin (v1.1.0): Added managing-hooks skill with 5 core workflows for hook creation, analysis, updating, plugin composition, and evaluation
- Comprehensive documentation of all 10 hook event types including prompt-based and plugin hooks
- 3 real-world examples: intelligent Stop hook, Python formatter plugin, MCP tool security patterns
- Complete coverage: All 10 hook events (PreToolUse, PermissionRequest, PostToolUse, SessionStart, SessionEnd, Stop, SubagentStop, UserPromptSubmit, Notification, PreCompact), command and prompt-based hooks, plugin merging behavior, environment variables (${CLAUDE_PLUGIN_ROOT}, ${CLAUDE_ENV_FILE}), MCP tool integration consolidated in configuration guide
- Progressive disclosure: 9 supporting files, 3 working examples, 3 production-ready templates
- 702-line SKILL.md (~4,914-7,020 tokens) with MCP patterns consolidated into configuration-guide.md
Full Changelog: v1.2.0.0...v1.3.0.0
v1.2.0.0 - Claude Code capabilities plugin
Created claude-code-capabilities plugin (v1.0.0) for managing Claude Code skills, commands, hooks, MCPs, prompts, and subagents
- Added managing-agent-skills skill with 4 core workflows for creating, analyzing, and improving agent skills
- Complete alignment with official Claude Code documentation
- 17 major capability areas: Technical Architecture, Security Considerations, API Integration, Package Dependencies, MCP Tool References, Skill Composition, Evaluation-Driven Development, Iterative Development, and more
- 10 supporting files with progressive disclosure architecture
- Token-based metrics and model-specific tuning recommendations
Full Changelog: v1.1.0.0...v1.2.0.0