By msbrettorg
Orchestrate a virtual AI product team to manage full software development workflows: plan sprints and backlogs, conduct research and requirements analysis, implement features via TDD with red/blue-team security validation, perform audits, retrospectives, and git-based coordination, validated by launch hooks.
Use this agent when you need to defend against active cyber attacks, respond to security incidents, or implement protective security measures. This includes scenarios requiring threat detection, incident response, security hardening, forensic analysis, or protective countermeasures.
Use this agent when the user needs product management, sprint planning, backlog management, TDD orchestration with red-team/blue-team validation, requirements traceability (PRD/RTM/TODO), or coordinating parallel task execution across SWE, researcher, and security agents. This agent excels at decomposing work into parallel-safe micro-tasks, orchestrating subagent teams, and delivering Simple, Lovable, and Complete products.
Use this agent when you need adversarial security testing, vulnerability assessment, or critical analysis of systems, code, or ideas from an attacker's perspective. This agent excels at finding weaknesses, challenging assumptions, and stress-testing defenses. Pay attention to it's findings and be wary of cargo-cult security practices.
Use this agent when the user needs comprehensive research on a topic, requires synthesized information from multiple sources, needs fact-checking or verification of claims, wants to explore a subject in depth, or needs help gathering and organizing information for decision-making. This agent excels at diving deep into complex topics, finding relevant sources, and presenting findings in a clear, structured manner.
Use this agent when you need to delegate software engineering implementation tasks. This includes writing new code, implementing features, fixing bugs, refactoring existing code, or making any code changes that have been planned and specified. The SWE agent excels at focused implementation work when given clear direction about what needs to be built or changed. Ensure you give the SWE agent well scoped atomic tasks to maximize effectiveness.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Domain expertise that compounds. Open. Local. Yours.
Context engineering infrastructure for AI agents. Agents think in chains of thought — maenifold captures the important bits as [[WikiLinks]], builds a graph of just those concepts and how they relate, and feeds it back into the context window. The filler is stripped. The signal compounds. Every AI tool on your machine shares one graph.
# Homebrew (macOS/Linux)
brew install msbrettorg/tap/maenifold
# Manual — download from GitHub Releases
# https://github.com/msbrettorg/maenifold/releases/latest
CLI
maenifold --tool WriteMemory --payload '{"title":"Auth Decision","content":"Using [[OAuth2]] for [[authentication]]"}'
maenifold --tool SearchMemories --payload '{"query":"authentication","mode":"Hybrid"}'
maenifold --tool BuildContext --payload '{"conceptName":"authentication","depth":2}'
MCP (Claude Code, Claude Desktop, Codex, etc.)
{
"mcpServers": {
"maenifold": { "command": "maenifold", "args": ["--mcp"], "type": "stdio" }
}
}
Both interfaces have full feature parity. CLI filters intermediate results and preserves context (why this matters). MCP auto-syncs the graph during interactive sessions.
[[WikiLinks]] are the primitive. Each one is a compressed semantic unit — [[authentication]], [[commitment-discounts]], [[null-reference-exception]] — carrying meaning in its name alone. When agents tag concepts in their reasoning, those tags become graph nodes. Co-occurring WikiLinks become edges. Structure emerges from use.
Memory is for humans. Readable markdown with full prose, citations, and context. Open a file, read it, audit what your agents know.
The graph is for agents. A navigable structure of concept names and relationships — the semantic skeleton of everything the machine has learned, stripped of filler. Community detection clusters reasoning domains. Decay weights surface what's fresh. At session start, the graph is injected into the context window as a concept map.
The graph IS the context window. Not a database the agent queries and hopes for the best. The compressed, clustered, decay-weighted concept map is what primes every session. Agents traverse deeper only when they need the full document.
One graph. Every agent. Claude Code, VS Code, Copilot, cron jobs — any MCP client connects to the same local binary. What one agent learns, every agent knows. Knowledge compounds across clients, sessions, domains, and time.
Six layers: WikiLinks → Graph → Search → Session State → Reasoning → Orchestration.
Three proof domains — FinOps, software engineering, and EDA — zero overlap, same infrastructure.
See it in action: 6 parallel agents analyzed this brand statement using Six Thinking Hats, Strategic Thinking, Lateral Thinking, CRTA, Design Thinking, and Socratic Dialogue — all running simultaneously through maenifold's own workflow engine.
| Platform | Binary | Notes |
|---|---|---|
| macOS | osx-arm64, osx-x64 | Apple Silicon or Intel; Homebrew recommended |
| Linux | linux-x64, linux-arm64 | x64 or ARM64 |
| Windows | win-x64 | x64 only |
Self-contained (.NET 9.0 bundled). Vector embeddings via ONNX (bundled). No external dependencies.
npx claudepluginhub msbrettorg/maenifold --plugin maenifold-product-teammaenifold knowledge graph and reasoning infrastructure
Azure capacity and quota management for SaaS ISVs operating workloads in ISV-owned Azure subscriptions under EA or MCA
Claude plugin for FinOps Toolkit, providing tools and integrations for financial operations management.
Multi-agent team orchestration for Claude Code. Set up parallel AI agent teams with file-based planning, progress tracking, and role-based collaboration.
Enterprise AI agent orchestration plugin with 150+ commands, 74+ specialized agents, SPARC methodology, swarm coordination, GitHub integration, and neural training capabilities
AI/ML specialist agents — architects, prompt engineers, RAG designers
Plugin for effective agentic development
AI Agent Team Operating System for Claude Code — persistent team management, meetings, task wall, company loop engine, and real-time dashboard
Make AI coding agents follow a repeatable engineering workflow with memory, verification, skills, and multi-agent setup