Meta

A two-stage system that captures user corrections during Claude sessions and queues them for review. Users run /reflect to apply learnings to CLAUDE.md files. It detects correction patterns like 'no, use X' and 'actually...' to help Claude remember preferences across sessions.

This skill enforces mandatory workflows for discovering and using other Claude skills. It requires checking for applicable skills before any task, reading them with the Skill tool, and announcing usage. The skill mandates TodoWrite for checklists and provides specific rationalizations to avoid.

Consolidates five separate agent management skills into one package for selecting, spawning, tracking, and processing outputs from Claude agents. Provides SQLite audit trails, dependency validation, and structured JSON outputs for multi-agent workflows.

Acts as a central dispatcher that analyzes user queries and routes them to appropriate domain-specific routers (startup, engineering, operations). Provides clear routing logic with keyword detection and decision trees. Handles 72+ specialized skills across three domains with example workflows.

A framework skill for creating and validating Claude skills with strict structural guidelines. Enforces TitleCase naming, flat folder hierarchies, and standardized YAML frontmatter. Provides step-by-step conversion procedures for migrating legacy skills to the canonical format.

A specialized agent for extracting insights from multi-agent interactions. It analyzes workflows, identifies patterns, and builds a knowledge graph to suggest improvements. It integrates with vector and graph databases to store and retrieve learnings.

A bash script that scans task descriptions for keywords and outputs comma-separated agent categories. Simple pattern matching helps Claude Flow Novice coordinate agent selection by suggesting frontend, backend, devops, etc. It's stateless and integrates with existing agent selector tools. Clear examples show common use cases.

This skill automates brand compliance checks across visual design, tone of voice, legal claims, and accessibility standards. It loads guidelines from config files, runs parallel validations, and generates detailed reports with specific fixes. It's useful for marketing teams and designers to ensure consistency.

Applies TDD methodology to skill testing by requiring baseline failure observation before skill creation. Provides concrete pressure scenarios combining time, authority, and sunk cost factors to verify skill compliance under realistic conditions. Includes rationalization tables to document and counter common excuses for rule violations.

Enforces mandatory skill usage protocol for Claude, requiring checking for relevant skills before any task. Creates TodoWrite todos for checklist items and announces skill usage transparently. Prevents rationalization of skipping established workflows.

This skill provides a structured framework for B2B lead generation, defining lead types, funnel stages, gating strategies, and attribution models. It focuses on operational execution with clear tables, decision trees, and diagnostic questions. Includes practical templates and separates core fundamentals from optional AI automation.