🔌
Plugins
Loading...
Showing 30 of 23036 results
Tool · Cost-free, Ollama-first code review — multi-angle finders, adversarial verify, structured verdict.
Harness · Autonomous coding loop over Claude Code + Kagura Memory (run / review / goal). Stateful, resumable, creates PRs, HITL gates — not an atomic tool.
Persistent AI project memory — restore session context, recall & save knowledge, setup, and an MCP smoke test
Substrate · Memory Cloud. Persistent AI memory grounding the whole ecosystem — session-start/session-summary bracket every session; recall/remember/guide used throughout.
Tool · Memory-grounded PLAN-layer over Claude Code + Kagura Memory. Thin skill wrapping the kagura-planner CLI (plan / doctor).
Claude Code plugin for sending notifications and alerts to Telegram.
Live KanBan dashboard for monitoring Claude Code agent teams in real-time
Autonomous, card-driven kanban development: an orchestrator and specialist agents that run each backlog card through slice → design → implement → test → review, shipping design and implementation as two reviewable PRs per card.
v2.13.0 (KBT-F470) completes app-scoped Version-CRUD on the plugin surface: adds /kanbantic-version-create (non-destructive wrapper around the app-scoped create_version) and /kanbantic-version-delete (mutating-confirm wrapper around the new delete_version tool, warning that delete cascades to the Version's associated issues — KBT-BD158), and re-syncs known-mcp-tools.json to add delete_version (4 legacy release-tools still excluded). v2.12.0 completes the version-aware track (KBT-E066 / F318-F320): release→Version rename across the lane-skills, known-mcp-tools.json curated to the live Version tools (12 in, 4 legacy release-tools out), lint Invariant 5 against stale release tokens, and two zero-dep hooks (locked-version-blocker on claim_issue + Stop version-summary). First Application self-released through the new Version-flow. KBT-E087 Agent-orchestratie & modelbeleid: /kanbantic-orchestrate sequencer skill (F436), launch-orchestrator script (F438), context-aware lane-skill gates for Cowork/no-repo (F447), and sync `model:` frontmatter for the new Subagent Model field (F437).
Claude gada do Ciebie jak Kapitan Tytus Bomba — oficer Gwiezdnej Floty z kreskowki Bartosza Walaszka: krzykliwie, wulgarnie i z kanonicznymi cytatami. Styl tylko w czacie; kod, commity i PR-y zostaja czyste.
System architecture skills: C4 model diagrams, ADR, architecture decision records, system design, GCP infrastructure, architecture-principles (modernization / Team Topologies / evolutionary architecture / DORA), Thoughtworks Tech Radar structural index + themes (volume-stamped)
UI/UX and visual design skills: Apple HIG, Material Design 3, icon design, frontend aesthetics, UX psychology, Hi-Fi design quality, mobile auth screen design, design system, Garrett UX analysis (Strategy/Scope/Structure/Skeleton/Surface 5-layer framework)
Code quality, review, and design doc skills: Codex-based code review with session resume, Google design docs, code refactoring, QA management, and Node.js 24 backend development (Express, Drizzle/Prisma, Vitest, tsup)
Meta-skills that observe the Claude Code session itself: Stop-hook gated learning-candidate recording, session-pointer breadcrumbs into auto-memory
Product management agents: requirements analysis, agile project management
Windows Desktop development skills: .NET 10 WPF/XAML, MVVM, Clean Architecture, Vertical Slice, MSIX packaging/signing/CI, Serilog diagnostics, EF Core migration (SQLite offline-safe), performance profiling, xUnit/Moq testing, dotnet-windows-developer agent
JDB (Java Debugger) for Claude Code via MCP - attach to running JVMs and debug with breakpoints, expression evaluation, thread inspection, and class introspection
LSP-backed code intelligence for Claude Code via MCP and native LSP - navigate definitions, references, call hierarchies, type hierarchies, and diagnostics across codebases
rr reverse debugging for Claude Code via MCP - record, replay, and debug program executions with full forward and reverse control
Track how your brand shows up in AI search across Claude, ChatGPT, Perplexity, and Gemini, and draft page-copy fixes for the gaps.
Active coding discipline enforcer based on Karpathy's 4 principles: surface assumptions, simplify, make surgical changes, define verifiable goals. Ships 4 Python tools (complexity_checker, diff_surgeon, assumption_linter, goal_verifier), a review agent, /karpathy-check command, and pre-commit hook. All stdlib-only.
Active coding discipline enforcer based on Karpathy's 4 principles: surface assumptions, simplify, make surgical changes, define verifiable goals. Ships 4 Python tools (complexity_checker, diff_surgeon, assumption_linter, goal_verifier), a review agent, /karpathy-check command, and pre-commit hook. All stdlib-only.
Karpathy-inspired behavioral rules for Claude Code (via ruler-engine). Requires ruler-engine 0.2.0+.
Karpenter node autoscaling expert for EKS — NodePools, EC2NodeClasses, EKS Auto Mode NodeClasses, consolidation/disruption tuning, spot adoption, GPU pools, cost optimization, and node-level troubleshooting
Progressive code disclosure and behavioral coupling analysis via LSP and git history