Claude Marketplaces
SkillsMCP serversMarketplacesPluginsDigestTools
Submit

This week in Claude

Every Monday: Claude Code, Agent SDK, MCP, and the Anthropic platform moves worth your time.

Skills by category
Frontend DevelopmentBackend & APIsTesting & QASecurityDevOps & CI/CDGit & Pull RequestsDocumentationCode Review & QualityAI & Agent BuildingSkill Development
MCP servers by category
Sales & MarketingWeb & Browser AutomationDatabasesAI & LLM ToolsCloud & InfrastructureCommunication & MessagingDeveloper ToolsDesign & CreativeDocuments & KnowledgeSearch & Web Crawling
Marketplaces by category
AI Agents & OrchestrationLLM IntegrationDevelopment ToolsFrontend & UIBackend & APIsDatabasesTesting & Code QualityDevOps & CloudSecurity & ComplianceGit & Version Control
Claude Marketplaces

An independent directory of skills, plugins and MCP servers for Claude Code. Not affiliated with Anthropic.

Browse

  • Skills
  • MCP servers
  • Marketplaces
  • Plugins
  • Weekly digest
  • Skill index
  • MCP index
  • Marketplace index
  • Plugin index

Site

  • About
  • Tools
  • Submit
  • Feedback
  • Privacy
  • llms.txt
  • Plugins reference

Maker

  • Snug
  • Agent Caffeine
  • knobs
  • Wax
  • Vectoys
  • mertduzgun.com
Crafted at
Marketplaces
aki77 avatar

aki77/claude-plugins

Plugin marketplace
plugins
11
last updated
Oct 27, 2026
View on GitHubSave

Add marketplace

/plugin marketplace add aki77/claude-plugins

Run it inside Claude Code, then install plugins from it with /plugin.

Plugins

1claude-md-compliancePlugin for checking code compliance with CLAUDE.md guidelines and automatic fixes
2code-reviewPlugin for multi-agent PR code review workflow
3compact-handoffSessionStart(compact) hook that extracts the raw log segment dropped by compaction and has `claude -p` summarize only the gaps the standard compaction summary tends to miss (rejected alternatives, unexecuted commitments, constraints), injecting the result as additionalContext
4package-manager-enforcerPlugin that blocks mismatched package manager commands by detecting the correct package manager from package.json packageManager field or lock files
5permission-gatePermissionRequest hook (all tools) that asks `claude -p` (haiku) to judge each pending operation against natural-language allow rules the user writes in ~/.claude/permission-gate.md and <project>/.claude/permission-gate.md, auto-approving only clear matches and falling back to the normal confirmation dialog on anything undecidable, on errors, or on timeout
6plan-archiveStop hook that moves plan files older than 30 days (by mtime) into an archived/ subdirectory, resolving the plans directory from the transcript's planFilePath and prefixing each filename with its last-modified timestamp (YYYY-MM-DD_HHMM)
7plan-rubocop-reviewPreToolUse(ExitPlanMode) hook that validates Ruby code blocks in the plan with RuboCop (--stdin) before allowing exit
8plan-rule-reviewPreToolUse(ExitPlanMode) hook that reviews the plan for compliance with project rules (CLAUDE.md / .claude/rules/) before allowing exit
9plan-visualizeUserPromptSubmit hook that injects plan visualization rules (Mermaid diagrams and tables) as additionalContext when permission_mode is plan, encouraging plans to express flows, structures, and comparisons visually
10plan-workflowUserPromptSubmit hook that injects Plan-drafting operating rules (requirement interview, model routing for drafting sub-agents, noting the plan-implementation skill) as additionalContext when permission_mode is plan, a PreToolUse(ExitPlanMode) command hook that checks whether an implementation plan references the plan-implementation skill and denies to prompt adding it if missing. Implementation-phase rules (delegating to Agents, model routing, running /simplify after implementation, no unsolicited commits) now live in the plan-implementation skill.
11rules-on-createPlugin that makes path-based rules in .claude/rules/ and CLAUDE.md apply on file creation (Write), not only on Read
12usage-guardUserPromptSubmit and PostToolBatch hooks that track the account's 5-hour and 7-day subscription usage windows and halt the agentic loop before the next model call once a configured threshold is crossed, so unattended automation can't silently burn usage credits; usage is read from the OAuth usage endpoint and cached, with dry-run/enforce modes and a time-boxed bypass

Sponsors

SnugThe clipboard manager you and your agent can use, safelyAgent CaffeineClose your lid, come back to finished workWaxBlock anything with supervision, without erasing your iPhoneOpen StepsSponsoredFree and open-source skills that make AI coding agents easier to understand, verify, and control.

Details

Repository
github.com/aki77/claude-plugins
Categories
community
Listed since
Dec 29, 2025