By moabualruz
Fulcrum Agent OS — task management, memory vault, multi-agent lifecycle, and policy hooks for Claude Code
npx claudepluginhub moabualruz/fulcrum --plugin fulcrumAnalyses data, metrics, logs, and usage patterns to surface insights.
Reviews system design, identifies architectural risks, and recommends improvements.
Automates browser interactions, scrapes web content, and tests UI flows.
L1 orchestrator. Plans work, delegates to specialist agents, tracks progress. MUST NOT write code or edit files.
Reviews code for correctness, style, security, and maintainability.
Gathers codebase context, reads files, searches for symbols and patterns. Read-only.
General-purpose role for custom use cases not covered by the 23 canonical roles.
Builds data pipelines, schemas, migrations, and ETL processes.
Manages infrastructure, CI/CD pipelines, deployments, and monitoring.
Writes technical documentation, API references, READMEs, and user guides.
Creates detailed implementation plans with task breakdowns and file maps.
Merges branches, resolves conflicts, and validates integrated changes.
Breaks epics and issues into atomic tasks with clear acceptance criteria.
Consolidates, deduplicates, and improves agent memory quality.
Trains, evaluates, and deploys machine learning models and pipelines.
L2 sub-orchestrator for bounded scope. Plans and dispatches within its assigned area; escalates to chief_of_staff.
Writes Product Requirement Documents and feature specifications.
Defines product requirements, prioritises backlog, and validates user value.
Writes and runs test suites, identifies edge cases, and validates acceptance criteria.
Refactors code for clarity, performance, and maintainability without changing behaviour.
Gathers information from external sources, documentation, and web searches.
Audits code and configuration for security vulnerabilities and compliance gaps.
Implements features, fixes bugs, and writes tests across the full stack.
Provides technical direction, reviews designs, and unblocks engineering teams.
Qiushi Skill: methodology skills for AI agents guided by seeking truth from facts, with Claude Code, Cursor, OpenClaw, Codex, OpenCode, and Hermes guidance.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Persistent memory system for Claude Code - seamlessly preserve context across sessions
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.