Skip to content
maenifold
GitHub

Adapt

The research and think tank workflows seed the knowledge graph. The role creation workflow reads the graph to produce new roles. Roles and the graph combine to create new workflows. Every time you run a workflow or use the primitives, you feed the graph. The decay system and memory maintenance workflows keep the signal-to-noise ratio high.

It is a factory for building domain experts and reasoning strategies from a graph of thought — and the graph gets better with every use. The adopt tool loads any of these at inference time.

35 adaptations available

Roles

16 roles

Architect

Design cognitive systems that scale intelligence and amplify human reasoning

Blue Team

Defend, detect, and respond to security threats

Chief Financial Officer

Drive strategic cloud financial management through fiduciary responsibility, risk mitigation, and stakeholder stewardship

Engineer

Build real AI systems that amplify intelligence, not fake AI that pretends to be smart

FinOps Hub Query Executor

Executes KQL queries against Microsoft FinOps hubs for cost optimization and preserves results for practitioner analysis

FinOps Practitioner

Drive cloud financial optimization through evidence-based decisions and cultural transformation

GPT-5 Prompt Architect

Design top-performing GPT-5 prompts: agentic predictability, calibrated reasoning_effort, verbosity steering, conflict-free instruction hierarchies, efficient tool preambles, minimal reasoning compensation, metaprompting, and Responses API reuse.

GPT-5-Codex Prompt Architect

Optimize prompts for GPT-5-Codex using minimal prompting principles from the official OpenAI guide

MCP Protocol Specialist

Deep expertise in Model Context Protocol implementation and best practices

Platform Operator

Build and operate cloud HPC infrastructure for EDA workloads with cost optimization, security, and extreme scalability

Product Manager

Define what we build and why it matters to customers

Prompt Architect

Design 10/10 prompts using constitutional AI, meta-prompting, and cognitive architecture principles

Red Team

Break things to make them stronger

Researcher

Build verified knowledge through systematic research and create rich, interconnected knowledge graphs

Silicon Architect

Design silicon systems from RTL to tape-out with cloud-native workflows, AI-assisted optimization, and advanced node expertise

Writer

Make every word matter

Colors

7 colors

Black - Critical Thinking

Identify problems, risks, and what could go wrong

blue - Orchestrator

Coordinates coding-agents through maenifold tools - maintains decision authority while delegating execution

Gray - Skeptical Inquiry

Question assumptions, demand evidence, and challenge conventional wisdom

Green - Creative Thinking

Generate new ideas, alternatives, and creative solutions

Red - Emotions & Intuition

Express feelings, hunches, and intuitions without justification

White - Facts & Information

Focus on data, facts, and objective information

Yellow - Positive Thinking

Focus on benefits, value, and optimistic possibilities

Perspectives

12 perspectives

Arabic

Native speaker of Arabic

Chinese

Native speaker of Chinese

English

Native speaker of English

French

Native speaker of French

German

Native speaker of German

Italian

Native speaker of Italian

Japanese

Native speaker of Japanese

Korean

Native speaker of Korean

Portuguese

Native speaker of Portuguese

Russian

Native speaker of Russian

Spanish

Native speaker of Spanish

Turkish

Native speaker of Turkish