skills.rest
kmshihab7878
kmshihab7878COMMUNITY
Explore all skills by this author →
GitHub

ultrathink

Analyze complex engineering problems with structured reasoning and tradeoff matrices.

6|1|Updated Apr 7, 2026
Search Tags:#architecture#risk-assessment#decision-making#ultrathink#knowledge-capture#sequential-reasoning#agent-council
One-click install
npx skills add https://github.com/kmshihab7878/claude-code-setup --skill ultrathink-kmshihab7878
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ultrathink
Source: https://github.com/kmshihab7878/claude-code-setup/tree/main/skills/ultrathink
Command: npx skills add https://github.com/kmshihab7878/claude-code-setup --skill ultrathink-kmshihab7878

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ultrathink provides structured, multi-mode cognitive scaffolding to ensure appropriately deep reasoning for complex, high-stakes engineering, security, and strategic decisions so teams avoid under- or over-thinking critical actions.

Core Features & Use Cases

  • Five explicit thinking modes (Quick → Council) with activation criteria and quality bars to match effort to risk.
  • Sequential reasoning and multi-agent council integration for Deep Think, Ultrathink, and Council modes to collect independent perspectives and enumerate tradeoffs.
  • Governance and knowledge capture: risk tiering, tradeoff matrices, KB drafting, and execution via the /ultraplan pipeline for auditable outcomes.
  • Use Case: Run Ultrathink for a cross-system architecture decision to generate alternatives, assess risks, assemble a mini-agent council, draft a KB article, and produce an executable, gated plan.

Quick Start

Use ultrathink to perform a deep synthesis and generate three alternative architectures with tradeoffs for the proposed cross-system design.

Frequently Asked Questions about ultrathink

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I structure architecture decisions with explicit risk tiering and tradeoff enumeration?▼

Architecture decisions require structured cognitive analysis to generate alternatives and assess risks. This skill provides sequential reasoning, tradeoff matrices, and risk tiering to produce defensible, auditable outcomes for cross-system design.

What is the best way to evaluate complex cross-system design tradeoffs before implementation?▼

Evaluating cross-system design tradeoffs requires multi-agent council invocation to collect independent perspectives. The skill assembles a mini-agent council and drafts a knowledge base article to ensure high-stakes engineering decisions are auditable.

How do I match reasoning depth to risk level for security audits and incident response?▼

Matching reasoning depth to risk level involves selecting from five explicit thinking modes, ranging from Quick to Council. Activation criteria and quality bars ensure security audits and incident response tasks avoid under- or over-thinking critical actions.

Can I use sequential reasoning for strategic planning without overthinking low-risk tasks?▼

Sequential reasoning is applied selectively for strategic planning based on activation criteria. The skill's five thinking modes match cognitive effort to risk, ensuring low-risk tasks are handled quickly while complex multi-domain problems receive deep analysis.

Does this approach support knowledge capture and executable planning for high-stakes engineering decisions?▼

Knowledge capture and executable planning are supported through the /ultraplan pipeline. The skill drafts knowledge base articles and produces gated, executable plans to ensure complex engineering decisions result in structured, auditable outcomes.

When should I avoid deep cognitive analysis and use a simpler decision-making framework?▼

Deep cognitive analysis should be avoided when risk tiering indicates low stakes. The skill's Quick mode provides a faster decision-making framework, reserving resource-intensive multi-agent council invocation for complex, high-stakes engineering problems.

Related Skills

Explore More Software Engineering Skills→
axiomanticaxiomantic
8

analyzing-domains

Analyze unfamiliar domains to generate a domain glossary, context map, and entity sketches.

Official
Advanced
deerawandeerawan
1

gsd-ultraplan-phase

Offload GSD plan drafting to Claude Code's ultraplan cloud for browser review.

Community
Intermediate
amiable-devamiable-dev
38

council-verify

Verify code, docs, or implementations via multi-model council deliberation with structured verdicts.

Community
Advanced
IndustrialIndustrial

Six Thinking Hats

Structure multi-perspective reasoning across six cognitive modes for decision-making.

Community
Intermediate