skills.rest
ruvnet
ruvnetCOMMUNITY
Explore all skills by this author →
Website𝕏@ruvGitHub

cognitive-pattern

Define and execute cognitive patterns for agent reasoning and decision-making workflows.

70.1k|8.4k|Updated Jun 2, 2025
Search Tags:#mcp#chain of thought#decision-making#Workflow#cognitive patterns#agent reasoning#Multi-Agent
One-click install
npx skills add https://github.com/ruvnet/claude-flow --skill cognitive-pattern
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cognitive-pattern
Source: https://github.com/ruvnet/claude-flow/tree/main/plugins/ruflo-daa/skills/cognitive-pattern
Command: npx skills add https://github.com/ruvnet/claude-flow --skill cognitive-pattern

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Agents often reason ad hoc, making it hard to codify, reuse, and improve decision-making strategies across tasks. This Skill lets you define structured cognitive patterns, embed them in executable workflows, and track how well they perform over time.

Core Features & Use Cases

  • Pattern Definition: Codify reasoning strategies such as decision trees, chain-of-thought decomposition, analogy mapping, and consensus evaluation.
  • Workflow Execution: Embed patterns into executable workflows and run them through the DAA (Decentralized Autonomous Agents) MCP tools.
  • Learning & Sharing: Evaluate pattern performance via learning status and propagate successful patterns through knowledge sharing.
  • Use Case: Define a consensus pattern where multiple agent perspectives are evaluated before a decision, execute it as a workflow, then check learning status to see if the pattern improves outcomes.

Quick Start

Define a chain-of-thought cognitive pattern for code review decisions, create a workflow around it, execute it, and report its learning status.

Frequently Asked Questions about cognitive-pattern

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

FAQPage Schema
How do I define a cognitive pattern for agent reasoning?▼

Call the daa_cognitive_pattern MCP tool with the pattern structure and its triggers. Then embed it in an executable workflow using daa_workflow_create and run it with daa_workflow_execute.

What types of cognitive patterns can agents use?▼

Four pattern types are supported: decision trees for structured if/then reasoning, chain of thought for step-by-step decomposition, analogy mapping for solving new problems from similar solved ones, and consensus patterns for multi-perspective evaluation.

How do I measure whether a cognitive pattern is working?▼

Call the daa_learning_status tool after executing the pattern's workflow. It reports how the pattern performs so you can decide whether to refine or share it.

Can cognitive patterns be shared between agents?▼

Yes, the daa_knowledge_share tool propagates successful patterns across agents. This lets proven reasoning strategies be reused instead of redefined for each agent.

When should I use a consensus pattern instead of chain of thought?▼

Use consensus patterns when a decision benefits from multiple independent perspectives before committing, such as high-stakes choices. Chain of thought fits problems that decompose cleanly into sequential reasoning steps.

Related Skills

Explore More Software Engineering Skills→
NVIDIA-NeMoNVIDIA-NeMo
1.8k

nemotron-add-pattern

Generate Nemotron step pattern files with frontmatter and validate via pytest.

Official
Intermediate
adarshmishra07adarshmishra07
18

mission-control

Manage AI agent workflows via a Kanban task board and real-time feed.

Community
Advanced
komlukkomluk
1

agent-memory

Manage a 3-tier persistent memory system for AI agents.

Community
Intermediate
CodeStorm-HubCodeStorm-Hub

saas-architecture-deep-dive

Explain and design scalable SaaS architectures with tenancy models and trade-offs.

Official
Advanced