falsify
13273Plugin Claude Code
Bundles 1 skill, 1 command · 143 tokens together
The scientific thinking protocol for AI agents. Falsify before you believe.
21,018 that install into Claude Code. Same mod, same page, whichever agent you run: a URL per tool would split one page into five that compete.
Plugin Claude Code
Bundles 1 skill, 1 command · 143 tokens together
The scientific thinking protocol for AI agents. Falsify before you believe.
Plugin Claude Code
Bundles 3 commands, 1 hook · 25 tokens together
Implementation of the Ralph Wiggum technique - continuous self-referential AI loops for interactive iterative development. Run Claude in a while-true loop with the same prompt until task completion.
Plugin Claude Code
Bundles 3 skills · 254 tokens together
Quy trình tạo Agent Skill chuẩn top-1.
Plugin Claude Code
Bundles 1 skill · 90 tokens together
Targeted codebase audit skill: parallel domain recon agents, adversarial false-positive elimination, machine-validated findings JSON, and a SPEC handoff for implementation.
Plugin Claude Code
Bundles 1 skill · 23 tokens together
Interact with Apple Mail on macOS. Read, send, search, and manage emails directly through Claude.
Plugin Claude Code
Bundles 3 skills, 1 hook · 371 tokens together
A requirements-clarification tool using three viewpoints: interview, unknowns, and the relationship between content and form.
childbamboo/claude-code-marketplace-sample
Plugin Claude Code
Bundles 6 skills, 1 command, 2 MCP servers · 248 tokens together
A general development add-on that combines MCP server integrations with skills for processing documentation. MCP is a way for an agent to use tools provided by another local or remote service.
MikSkrzyp/identity-storage-mcp
Plugin Claude Code
Portable, auditable long-term memory for AI agents. Backed by local SQLite. Search past memories, store new ones with automatic classification.
Plugin Claude Code
Local, cross-agent memory for learning from real failures without treating requirement changes as mistakes.
Plugin Claude Code
Bundles 1 skill, 2 hooks · 108 tokens together
A plugin built on the Meshy MCP service that keeps local copies of models and provides a record book, full-text search, a web gallery, and warnings for models that were not saved. Its three-day expiry protection helps address temporary availability.
Plugin Claude Code
Query and mutate Neo4j graph databases from Claude Code via MCP. Each project supplies its own credentials through a local .env file.
hjsh200219/trading-by-clinical-trial
Plugin Claude Code
Bundles 6 skills, 3 agents, 1 MCP server · 1,076 tokens together
A plugin for analysing Korean pharmaceutical and biotechnology stocks using clinical-trial data from ClinicalTrials.gov and public market data from Naver Finance. It assigns a score out of 100, decision labels, and technical indicators without requiring the KIS API.
Plugin Claude Code
Bundles 7 skills · 380 tokens together
Expert skills for building n8n workflows with n8n-mcp.
Plugin Claude Code
Bundles 2 skills, 5 commands, 1 MCP server · 130 tokens together
Execute terminal commands on the host machine with configurable permissions. Provides secure shell access for developers with allowlist-based command filtering, directory restrictions, and multiple permission modes.
Plugin Claude Code
Bundles 1 skill · 0 tokens together
Workflow for managing feature spec and orchestating devteam delegations.
Plugin Claude Code
Standalone semantic-memory MCP server (SQLite + sqlite-vec + Ollama embedder). Stores and recalls memories by meaning; persona-agnostic.
Plugin Claude Code
Bundles 4 skills, 1 MCP server · 149 tokens together
Web search, content extraction, and media download.
shutovdef-dotcom/google-search-console-mcp
Plugin Claude Code
Bundles 4 skills, 1 MCP server · 129 tokens together
Connect Google Search Console to Claude Code. Query SEO data, inspect URLs, audit indexing, and identify content opportunities through natural language.
Plugin Claude Code
Bundles 1 skill, 1 MCP server · 60 tokens together
Run learning as a verified loop: build a detailed Target, then teach → lab → test → gate per claim, graded by a fresh-context examiner subagent.
Plugin Claude Code
Broadcast bus for peer Claude Code sessions: a broadcast() MCP tool to send and a UserPromptSubmit hook to receive, over JSONL mailboxes scoped to one project, a group of projects, or the whole machine.
Plugin Claude Code
Bundles 25 skills, 15 agents, 7 hooks, 1 MCP server · 4,063 tokens together
Amplificador consciente del modelo sobre Claude Code: rutea Fable/Opus/Sonnet/Haiku por dificultad de subtarea (auto-inyecta model=), mide costo/disciplina de routing, adapta el recall al modelo, y guarda su propia memoria (texto propio, desacoplado de claude-mem). 24 hooks (7 eventos), 7 MCP tools, guards 2…
yugmarwaha/todoist-weekly-review-mcp
Plugin Claude Code
Bundles 1 command · 15 tokens together
Human-in-the-loop weekly review for Todoist: surfaces overdue and stale tasks with signals, proposes fixes (including splitting hidden projects), applies only what you approve. Unofficial — not affiliated with Doist.
Plugin Claude Code
Bundles 1 skill · 58 tokens together
Drive a live ArcGIS Pro session from Claude via arcpy. Exposes executearcpy, listlayers, describe, projectinfo, and listener controls, with credit-gating on Esri credit-consuming operations. Requires ArcGIS Pro installed and open on the same Windows machine.
davekim917/design-artifact-loop
Plugin Claude Code
Bundles 1 skill, 1 MCP server · 124 tokens together
Render-grounded, non-slop UI design artifacts: commit a concrete design system before markup, then iterate against a deterministic linter + an independent vision critic, bounded by a round-capped state machine.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: