GitHub 3.6K

See how patterns cluster, overlap, and pull on each other.

Switch between shared tags, explicit links, and similarity views to understand which patterns reinforce each other before you decide what to build.

Showing 154 of 154 patterns · 0 connections · Shared tags
Abstracted Code Representation...
Abstracted Code Representation for Review UX & Collaboration proposed
Action Caching & Replay Patter...
Action Caching & Replay Pattern Reliability & Eval emerging
Action-Selector Pattern
Action-Selector Pattern Orchestration & Control emerging
Adaptive Sandbox Fan-Out Contr...
Adaptive Sandbox Fan-Out Controller Reliability & Eval emerging
Agent-Assisted Scaffolding
Agent-Assisted Scaffolding UX & Collaboration validated-in-production
Agent-Driven Research
Agent-Driven Research Orchestration & Control established
Agent-First Tooling and Loggin...
Agent-First Tooling and Logging Tool Use & Environment established
Agent-Friendly Workflow Design
Agent-Friendly Workflow Design UX & Collaboration best-practice
Agent Modes by Model Personali...
Agent Modes by Model Personality Orchestration & Control emerging
Agent-Powered Codebase Q&A / O...
Agent-Powered Codebase Q&A / Onboarding Context & Memory validated-in-production
Agent Reinforcement Fine-Tunin...
Agent Reinforcement Fine-Tuning (Agent RFT) Learning & Adaptation emerging Train model weights end-to-end on agentic tasks via reinforcement learning with real tool calls and custom reward signals, optimizing for domain-specific tool use efficiency and multi-step reasoning performance.
Agent SDK for Programmatic Con...
Agent SDK for Programmatic Control Tool Use & Environment emerging
Milestone Escrow for Agent Res...
Milestone Escrow for Agent Resource Funding UX & Collaboration emerging
Agentic Search Over Vector Emb...
Agentic Search Over Vector Embeddings Tool Use & Environment best-practice TODO: Add a concise summary for "Agentic Search Over Vector Embeddings" describing the pattern's purpose and key benefits.
AI-Accelerated Learning and Sk...
AI-Accelerated Learning and Skill Development UX & Collaboration validated-in-production
AI-Assisted Code Review / Veri...
AI-Assisted Code Review / Verification Feedback Loops emerging
AI Web Search Agent Loop
AI Web Search Agent Loop Tool Use & Environment emerging
Anti-Reward-Hacking Grader Des...
Anti-Reward-Hacking Grader Design Reliability & Eval emerging Design reward functions with multi-criteria evaluation and iterative hardening to prevent models from gaming graders, ensuring training rewards align with actual task quality.
Asynchronous Coding Agent Pipe...
Asynchronous Coding Agent Pipeline Reliability & Eval proposed
Autonomous Workflow Agent Arch...
Autonomous Workflow Agent Architecture Orchestration & Control established TODO: Add a concise summary for "Autonomous Workflow Agent Architecture" describing the pattern's purpose and key benefits.
Background Agent with CI Feedb...
Background Agent with CI Feedback Feedback Loops validated-in-production ## Problem Long-running tasks tie up the editor and require developers to babysit the agent.
Budget-Aware Model Routing wit...
Budget-Aware Model Routing with Hard Cost Caps Orchestration & Control established
Burn the Boats
Burn the Boats Orchestration & Control emerging
Canary Rollout and Automatic R...
Canary Rollout and Automatic Rollback for Agent Policy Changes Reliability & Eval established
Chain-of-Thought Monitoring & ...
Chain-of-Thought Monitoring & Interruption UX & Collaboration emerging Implement active surveillance of agent reasoning with capability to interrupt and redirect before completing flawed execution sequences, preventing wasted time on fundamentally wrong approaches.
CLI-First Skill Design
CLI-First Skill Design Tool Use & Environment emerging Design all skills as CLI tools first for dual-use by humans and agents, enabling manual debugging, programmatic invocation, composition with Unix tools, and transparent shell-based execution without building separate interfaces.
CLI-Native Agent Orchestration
CLI-Native Agent Orchestration Tool Use & Environment proposed
Code Mode MCP Tool Interface I...
Code Mode MCP Tool Interface Improvement Pattern Tool Use & Environment established LLMs generate TypeScript code to orchestrate MCP tools in ephemeral V8 isolates, eliminating token-heavy round-trips and enabling efficient multi-step workflows with 10x+ token savings.
Code-Over-API Pattern
Code-Over-API Pattern Tool Use & Environment established Agents write and execute code that processes data in execution environment instead of making direct API calls, dramatically reducing token consumption by keeping intermediate data out of context window (150K → 2K tokens).
Code-Then-Execute Pattern
Code-Then-Execute Pattern Tool Use & Environment emerging
Codebase Optimization for Agen...
Codebase Optimization for Agents UX & Collaboration emerging
Coding Agent CI Feedback Loop
Coding Agent CI Feedback Loop Feedback Loops best-practice
Compounding Engineering Patter...
Compounding Engineering Pattern Learning & Adaptation emerging Codify all learnings from each feature into reusable prompts, slash commands, subagents, and hooks—making each feature easier to build by creating increasingly "self-teaching" codebase that accelerates productivity over time.
Context-Minimization Pattern
Context-Minimization Pattern Context & Memory emerging
Context Window Anxiety Managem...
Context Window Anxiety Management Context & Memory emerging
Context Window Auto-Compaction
Context Window Auto-Compaction Context & Memory validated-in-production
Continuous Autonomous Task Loo...
Continuous Autonomous Task Loop Pattern Orchestration & Control established
CriticGPT-Style Code Review
CriticGPT-Style Code Review Reliability & Eval validated-in-production
Cross-Cycle Consensus Relay
Cross-Cycle Consensus Relay Orchestration & Control emerging
Curated Code Context Window
Curated Code Context Window Context & Memory validated-in-production
Curated File Context Window
Curated File Context Window Context & Memory best-practice
Custom Sandboxed Background Ag...
Custom Sandboxed Background Agent Orchestration & Control emerging
Democratization of Tooling via...
Democratization of Tooling via Agents UX & Collaboration emerging
Deterministic Security Scannin...
Deterministic Security Scanning Build Loop Security & Safety proposed
Dev Tooling Assumptions Reset
Dev Tooling Assumptions Reset UX & Collaboration emerging
Discrete Phase Separation
Discrete Phase Separation Orchestration & Control emerging TODO: Add a concise summary for "Discrete Phase Separation" describing the pattern's purpose and key benefits.
Disposable Scaffolding Over Du...
Disposable Scaffolding Over Durable Features Orchestration & Control best-practice
Distributed Execution with Clo...
Distributed Execution with Cloud Workers Orchestration & Control emerging TODO: Add a concise summary for "Distributed Execution with Cloud Workers" describing the pattern's purpose and key benefits.
Dogfooding with Rapid Iteratio...
Dogfooding with Rapid Iteration for Agent Improvement Feedback Loops best-practice ## Problem Developing effective AI agents requires understanding real-world usage and quickly identifying areas for improvement. External feedback loops can be slow, and simulated environments may not capture all nuances.
Dual LLM Pattern
Dual LLM Pattern Orchestration & Control emerging
Dual-Use Tool Design
Dual-Use Tool Design Tool Use & Environment best-practice TODO: Add a concise summary for "Dual-Use Tool Design" describing the pattern's purpose and key benefits.
Dynamic Code Injection (On-Dem...
Dynamic Code Injection (On-Demand File Fetch) Tool Use & Environment established
Dynamic Context Injection
Dynamic Context Injection Context & Memory established
Economic Value Signaling in Mu...
Economic Value Signaling in Multi-Agent Networks Orchestration & Control experimental-but-awesome
Egress Lockdown (No-Exfiltrati...
Egress Lockdown (No-Exfiltration Channel) Tool Use & Environment established
Episodic Memory Retrieval & In...
Episodic Memory Retrieval & Injection Context & Memory validated-in-production
Explicit Posterior-Sampling Pl...
Explicit Posterior-Sampling Planner Orchestration & Control emerging
Extended Coherence Work Sessio...
Extended Coherence Work Sessions Reliability & Eval rapidly-improving
External Credential Sync
External Credential Sync Security & Safety validated-in-production
Factory over Assistant
Factory over Assistant Orchestration & Control validated-in-production
Failover-Aware Model Fallback
Failover-Aware Model Fallback Reliability & Eval validated-in-production
Feature List as Immutable Cont...
Feature List as Immutable Contract Orchestration & Control emerging TODO: Add a concise summary for "Feature List as Immutable Contract" describing the pattern's purpose and key benefits.
Filesystem-Based Agent State
Filesystem-Based Agent State Context & Memory established Agents persist intermediate results and working state to files, creating durable checkpoints that enable workflow resumption, recovery from failures, and support for long-running tasks.
Frontier-Focused Development
Frontier-Focused Development Learning & Adaptation emerging
Graph of Thoughts (GoT)
Graph of Thoughts (GoT) Feedback Loops emerging
Hook-Based Safety Guard Rails ...
Hook-Based Safety Guard Rails for Autonomous Code Agents Security & Safety validated-in-production
Human-in-the-Loop Approval Fra...
Human-in-the-Loop Approval Framework UX & Collaboration validated-in-production Systematically insert human approval gates for designated high-risk functions while maintaining agent autonomy for safe operations, with multi-channel approval interfaces and comprehensive audit trails.
Hybrid LLM/Code Workflow Coord...
Hybrid LLM/Code Workflow Coordinator Orchestration & Control proposed Configurable coordinator supporting both LLM-driven (flexible, fast iteration) and code-driven (deterministic, code review) workflows, enabling progressive enhancement from prototype to production-ready systems.
Incident-to-Eval Synthesis
Incident-to-Eval Synthesis Feedback Loops emerging
Inference-Healed Code Review R...
Inference-Healed Code Review Reward Feedback Loops proposed
Inference-Time Scaling
Inference-Time Scaling Orchestration & Control emerging
Initializer-Maintainer Dual Ag...
Initializer-Maintainer Dual Agent Architecture Orchestration & Control validated-in-production
Intelligent Bash Tool Executio...
Intelligent Bash Tool Execution Tool Use & Environment validated-in-production
Inversion of Control
Inversion of Control Orchestration & Control validated-in-production Traditional "prompt-as-puppeteer" workflows force humans to spell out every step, limiting scale and creativity.
Isolated VM per RL Rollout
Isolated VM per RL Rollout Security & Safety emerging Spin up an isolated virtual machine for each RL rollout to prevent cross-contamination between parallel agent executions, ensuring safe training with destructive tool access.
Iterative Multi-Agent Brainsto...
Iterative Multi-Agent Brainstorming Orchestration & Control experimental-but-awesome
Iterative Prompt & Skill Refin...
Iterative Prompt & Skill Refinement Feedback Loops established
Lane-Based Execution Queueing
Lane-Based Execution Queueing Orchestration & Control validated-in-production
Language Agent Tree Search (LA...
Language Agent Tree Search (LATS) Orchestration & Control emerging
Latent Demand Product Discover...
Latent Demand Product Discovery UX & Collaboration best-practice TODO: Add a concise summary for "Latent Demand Product Discovery" describing the pattern's purpose and key benefits.
Layered Configuration Context
Layered Configuration Context Context & Memory established
Lethal Trifecta Threat Model
Lethal Trifecta Threat Model Reliability & Eval best-practice
LLM-Friendly API Design
LLM-Friendly API Design Tool Use & Environment emerging
LLM Map-Reduce Pattern
LLM Map-Reduce Pattern Orchestration & Control emerging
LLM Observability
LLM Observability Reliability & Eval proposed Integrate LLM observability platforms for span-level tracing of agent workflows, providing visual UI debugging, workflow linking, and aggregate metrics to enable fast navigation of complex multi-step executions.
Memory Reinforcement Learning ...
Memory Reinforcement Learning (MemRL) Learning & Adaptation proposed TODO: Add a concise summary for "Memory Reinforcement Learning (MemRL)" describing the pattern's purpose and key benefits.
Memory Synthesis from Executio...
Memory Synthesis from Execution Logs Context & Memory emerging TODO: Add a concise summary for "Memory Synthesis from Execution Logs" describing the pattern's purpose and key benefits.
Merged Code + Language Skill M...
Merged Code + Language Skill Model Reliability & Eval emerging
Multi-Model Orchestration for ...
Multi-Model Orchestration for Complex Edits Orchestration & Control validated-in-production
Multi-Platform Communication A...
Multi-Platform Communication Aggregation Tool Use & Environment emerging Create unified search interface that queries all communication platforms in parallel and aggregates results into consistent format, enabling single-query cross-platform search with minimal latency through parallel execution.
Multi-Platform Webhook Trigger...
Multi-Platform Webhook Triggers Tool Use & Environment emerging Implement multi-platform webhook triggers (Notion, Slack, Jira, reacji, scheduled events) to allow external SaaS tools to automatically initiate agent workflows, enabling low-friction, reactive automation from existing platforms.
Artifact-Driven Analysis Pipel...
Artifact-Driven Analysis Pipeline Orchestration Orchestration & Control emerging
No-Token-Limit Magic
No-Token-Limit Magic Reliability & Eval experimental-but-awesome
Non-Custodial Spending Control...
Non-Custodial Spending Controls Security & Safety emerging
Opponent Processor / Multi-Age...
Opponent Processor / Multi-Agent Debate Pattern Orchestration & Control emerging TODO: Add a concise summary for "Opponent Processor / Multi-Agent Debate Pattern" describing the pattern's purpose and key benefits.
Oracle and Worker Multi-Model ...
Oracle and Worker Multi-Model Approach Orchestration & Control emerging
Parallel Tool Call Learning
Parallel Tool Call Learning Orchestration & Control emerging TODO: Add a concise summary for "Parallel Tool Call Learning" describing the pattern's purpose and key benefits.
Conditional Parallel Tool Exec...
Conditional Parallel Tool Execution Orchestration & Control validated-in-production
Patch Steering via Prompted To...
Patch Steering via Prompted Tool Selection Tool Use & Environment best-practice
PII Tokenization
PII Tokenization Security & Safety established Implement interception layer in MCP client that automatically tokenizes PII before reaching model and untokenizes for tool calls, enabling agents to orchestrate sensitive workflows without exposing raw data to LLM.
Plan-Then-Execute Pattern
Plan-Then-Execute Pattern Orchestration & Control established
Planner-Worker Separation for ...
Planner-Worker Separation for Long-Running Agents Orchestration & Control emerging
Proactive Agent State External...
Proactive Agent State Externalization Context & Memory emerging Modern models like Claude Sonnet 4.5 proactively externalize state through self-generated notes—enhanced through guided frameworks, hybrid memory architecture, and progressive state building to capture decision rationale and knowledge gaps.
Proactive Trigger Vocabulary
Proactive Trigger Vocabulary UX & Collaboration emerging TODO: Add a concise summary for "Proactive Trigger Vocabulary" describing the pattern's purpose and key benefits.
Progressive Autonomy with Mode...
Progressive Autonomy with Model Evolution Orchestration & Control best-practice TODO: Add a concise summary for "Progressive Autonomy with Model Evolution" describing the pattern's purpose and key benefits.
Progressive Complexity Escalat...
Progressive Complexity Escalation Orchestration & Control emerging Start agents with low-complexity, high-reliability tasks and progressively unlock more complex capabilities as models improve and trust is established, matching task complexity to current model capabilities for risk mitigation.
Progressive Disclosure for Lar...
Progressive Disclosure for Large Files Context & Memory emerging TODO: Add a concise summary for "Progressive Disclosure for Large Files" describing the pattern's purpose and key benefits.
Progressive Tool Discovery
Progressive Tool Discovery Tool Use & Environment established TODO: Add a concise summary for "Progressive Tool Discovery" describing the pattern's purpose and key benefits.
Prompt Caching via Exact Prefi...
Prompt Caching via Exact Prefix Preservation Context & Memory emerging
Recursive Best-of-N Delegation
Recursive Best-of-N Delegation Orchestration & Control emerging
Reflection Loop
Reflection Loop Feedback Loops established Generative models may produce subpar output if they never review or critique their own work.
Rich Feedback Loops > Perfect ...
Rich Feedback Loops > Perfect Prompts Feedback Loops validated-in-production
RLAIF (Reinforcement Learning ...
RLAIF (Reinforcement Learning from AI Feedback) Reliability & Eval emerging
Sandboxed Tool Authorization
Sandboxed Tool Authorization Security & Safety validated-in-production
Schema Validation Retry with C...
Schema Validation Retry with Cross-Step Learning Reliability & Eval emerging
Seamless Background-to-Foregro...
Seamless Background-to-Foreground Handoff UX & Collaboration emerging
Self-Critique Evaluator Loop
Self-Critique Evaluator Loop Feedback Loops established
Self-Discover: LLM Self-Compos...
Self-Discover: LLM Self-Composed Reasoning Structures Feedback Loops emerging
Self-Identity Accumulation
Self-Identity Accumulation Context & Memory emerging
Self-Rewriting Meta-Prompt Loo...
Self-Rewriting Meta-Prompt Loop Orchestration & Control emerging
Semantic Context Filtering Pat...
Semantic Context Filtering Pattern Context & Memory emerging
Shell Command Contextualizatio...
Shell Command Contextualization Tool Use & Environment established
Shipping as Research
Shipping as Research Learning & Adaptation emerging
Skill Library Evolution
Skill Library Evolution Learning & Adaptation established Agents persist working code implementations as reusable skills that evolve into well-documented capabilities over time, building organizational knowledge and reducing redundant problem-solving across sessions.
Soulbound Identity Verificatio...
Soulbound Identity Verification Security & Safety emerging
Spec-As-Test Feedback Loop
Spec-As-Test Feedback Loop Feedback Loops emerging
Specification-Driven Agent Dev...
Specification-Driven Agent Development Orchestration & Control proposed
Spectrum of Control / Blended ...
Spectrum of Control / Blended Initiative UX & Collaboration validated-in-production
Static Service Manifest for Ag...
Static Service Manifest for Agents Tool Use & Environment emerging
Stop Hook Auto-Continue Patter...
Stop Hook Auto-Continue Pattern Orchestration & Control emerging
Structured Output Specificatio...
Structured Output Specification Reliability & Eval established Constrain agent outputs using deterministic schemas that enforce structured, machine-readable results, enabling reliable validation, parsing, and integration with downstream systems.
Sub-Agent Spawning
Sub-Agent Spawning Orchestration & Control validated-in-production
Subagent Compilation Checker
Subagent Compilation Checker Reliability & Eval emerging
Subject Hygiene for Task Deleg...
Subject Hygiene for Task Delegation Orchestration & Control validated-in-production
Swarm Migration Pattern
Swarm Migration Pattern Orchestration & Control validated-in-production Main agent orchestrates 10+ parallel subagents working simultaneously on independent migration chunks, achieving 10x+ speedup for large-scale framework upgrades, lint rule rollouts, and API migrations.
Team-Shared Agent Configuratio...
Team-Shared Agent Configuration as Code UX & Collaboration best-practice Check agent configuration into version control as code, enabling consistent behavior across teams, faster onboarding, and collaborative improvement through PRs and code review.
Three-Stage Perception Archite...
Three-Stage Perception Architecture Orchestration & Control established
Tool Capability Compartmentali...
Tool Capability Compartmentalization Orchestration & Control emerging
Tool Search Lazy Loading
Tool Search Lazy Loading Context & Memory emerging Dynamically load tools via search instead of preloading all available tools to reduce context usage
Tool Selection Guide
Tool Selection Guide Orchestration & Control emerging
Tool Use Incentivization via R...
Tool Use Incentivization via Reward Shaping Feedback Loops emerging
Tool Use Steering via Promptin...
Tool Use Steering via Prompting Tool Use & Environment best-practice
Transitive Vouch-Chain Trust
Transitive Vouch-Chain Trust Security & Safety emerging
Tree-of-Thought Reasoning
Tree-of-Thought Reasoning Orchestration & Control established Linear chain-of-thought reasoning can get stuck on complex problems, missing alternative approaches or failing to backtrack.
Variance-Based RL Sample Selec...
Variance-Based RL Sample Selection Learning & Adaptation validated-in-production
Verbose Reasoning Transparency
Verbose Reasoning Transparency UX & Collaboration best-practice
Versioned Constitution Governa...
Versioned Constitution Governance Reliability & Eval emerging
Virtual Machine Operator Agent
Virtual Machine Operator Agent Tool Use & Environment established
Visual AI Multimodal Integrati...
Visual AI Multimodal Integration Tool Use & Environment emerging
Reliability Problem Map Checkl...
Reliability Problem Map Checklist for RAG and Agents Reliability & Eval proposed
Workflow Evals with Mocked Too...
Workflow Evals with Mocked Tools Reliability & Eval emerging TODO: Add a concise summary for "Workflow Evals with Mocked Tools" describing the pattern's purpose and key benefits.
Working Memory via TodoWrite
Working Memory via TodoWrite Context & Memory emerging
Workspace-Native Multi-Agent O...
Workspace-Native Multi-Agent Orchestration Orchestration & Control emerging
Zero-Trust Agent Mesh
Zero-Trust Agent Mesh Security & Safety established