MAJD DERBALI
Community@derbalimajd04-dot
MAJD DERBALI maintains a Codex-Flow skill registry covering multi-agent swarm orchestration, AgentDB vector memory, SPARC methodology, and GitHub CI/CD coordination.
Agent Skills by MAJD DERBALI
Showing 46 vetted skills indexed across 1 GitHub repositories.
V3 MCP Optimization
Implements connection pooling, load balancing, and tool registry optimization for MCP servers.
stream-chain
Chains sequential prompts so each step's output feeds the next workflow step.
sparc-methodology
Orchestrates multi-agent software development using the SPARC phased methodology with TDD workflows.
source-command-claude-flow-help
Displays Codex-Flow CLI commands and usage reference for agent orchestration.
Hooks Automation
Automates pre/post-operation hooks for agent coordination, formatting, and session memory.
V3 CLI Modernization
Modernizes Codex-flow v3 CLI with modular commands, interactive prompts, and hooks integration.
source-command-sparc-integration
Merges outputs from multiple development modes into a tested, cohesive system.
github-workflow-automation
Automates GitHub Actions workflows, CI/CD pipelines, and repository management with swarm coordination.
Verification & Quality Assurance
Verify code quality with truth scoring, automated checks, and git-based rollback.
source-command-sparc-devops
Automates infrastructure deployment, CI/CD pipelines, and cloud resource provisioning workflows.
V3 Deep Integration
Migrates Codex-flow onto agentic-flow@alpha adapters to eliminate duplicate orchestration code.
Skill Builder
Create Codex Skills with YAML frontmatter and progressive disclosure structure.
source-command-sparc-spec-pseudocode
Generates modular pseudocode specification files with TDD anchors from project requirements.
source-command-sparc-code
Generates modular code from pseudocode and architecture using the SPARC code mode.
V3 Core Implementation
Implements DDD domains and clean architecture patterns for TypeScript core modules.
ReasoningBank with AgentDB
Implement adaptive agent learning with trajectory tracking and vector-based memory retrieval.
source-command-claude-flow-memory
Store, query, and manage persistent memory entries in the Codex-Flow memory system.
ReasoningBank Intelligence
Implement adaptive learning for AI agents using ReasoningBank pattern recognition and strategy optimization.
V3 Security Overhaul
Remediates critical CVEs and implements secure-by-default patterns for Codex-flow v3.
swarm-advanced
Orchestrates multi-agent swarms for research, development, testing, and analysis workflows.
source-command-sparc-refinement-optimization-mode
Refactors code, enforces file size limits, and optimizes system performance via SPARC mode.
github-release-management
Orchestrates GitHub releases with automated versioning, testing, deployment, and rollback workflows.
AgentDB Advanced Features
Configure QUIC synchronization, hybrid search, and multi-database management for AgentDB vector stores.
browser
Automates web browser navigation, interaction, and data extraction using AI-optimized accessibility snapshots.
Frequently Asked Questions About MAJD DERBALI
FAQPage SchemaWhat tasks can I accomplish with MAJD DERBALI's Codex-Flow skills?▼
You can orchestrate multi-agent swarms for parallel task execution, implement semantic vector search and persistent agent memory with AgentDB, automate GitHub code review, releases, and multi-repo synchronization, apply the SPARC development methodology, and run AI-assisted pair programming with truth-score verification.
Who are these skills designed for?▼
These skills target developers and platform engineers building distributed multi-agent systems, RAG pipelines, and self-learning agents. They also serve DevOps engineers managing GitHub CI/CD pipelines, release orchestration, and project boards, plus teams adopting the SPARC specification-to-completion methodology.
What are the installation prerequisites and runtime dependencies?▼
GitHub skills require the GitHub CLI (gh) authenticated, git, Node.js v16+ (v20+ for release management), and Codex-flow or ruv-swarm configured. AgentDB skills need the AgentDB vector database. Browser automation uses Codex-flow browser tools such as open, snapshot, click, fill, and screenshot.
How does AgentDB improve agent memory and search performance?▼
AgentDB provides HNSW indexing delivering 150x to 12,500x faster search, quantization for 4-32x memory reduction, QUIC synchronization, and hybrid search. It supports session memory, long-term storage, pattern learning, and experience replay for stateful agents, chat systems, and RAG knowledge bases.
What does the SPARC methodology skill set include?▼
SPARC (Specification, Pseudocode, Architecture, Refinement, Completion) provides orchestrated modes including specification writer, auto-coder, debugger, security reviewer, optimizer, DevOps deployer, documentation writer, and post-deployment monitor, coordinated by a SPARC orchestrator with TDD and multi-agent delegation.